Individuals with syndromes, like Turner syndrome (TS), have been reported to have reduced growth hormone (GH) responsiveness - partially overcome by higher rhGH dosage (1). Moreover, the growth response in short children receiving recombinant human (rh)GH is highly variable independent on diagnosis but dependent on the individual GH responsiveness, which can be calculated based on the growth response as well as IGFI response on rhGH (2, 3).

Due to the phenotypical resemblance, NS has previously been called “male Turner syndrome (TS)” (4), despite different genetic background. Both syndromes are characterised by short stature, pterygium colli, webbed neck, cardiac anomalies, pectus deformities, and mild hearing loss. Usually, NS males suffer from cryptorchidism, while TS girls most often fail to enter puberty and are infertile (5, 6).

One of the most predominant features of children with NS and TS is short stature affecting more than 80% (7). Possible answers to the cause of poor growth in patients with NS include disturbances in GH/insulin-like growth factor-I (GH/IGFI) axis. Findings are however contradictory and both low (8, 9), normal (10) or high (11, 12) spontaneous GH secretion has been reported, as well as either low (13–16) or normal (17, 18) GH_max levels in response to stimulatory test. Still, all of these studies found lower level of IGFI.

So far, only two studies have displayed detailed analyses pattern of the spontaneous GH secretion and its correlation with growth in NS patients treated with rhGH. Both Ahmed et al. (11) with a pilot study of four patients, and Noordam et al. (8) with 16 prepubertal NS patients, showed a disorganised pattern categorized by irregular wide pulses and high trough concentrations in overnight GH secretion profiles particularly in boys with NS. However, despite the evidence for disturbed GH/IGFI axis, the patients responded to rhGH treatment, showing an increase in both height velocity and in IGFI level during one year of treatment.

Our overall aim was to investigate GH/IGF status, response in growth and IGFI, and responsiveness to rhGH treatment in children with NS compared to TS. In this paper, we present unique and detailed data on the 24hGH profile in 25 prepubertal children with NS, and we compare these data to that of girls with TS and to healthy normally growing children as a control group. Both growth response and response for IGFI and IGFBP3 during the first year of GH treatment were measured and compared. Our hypothesis was that children with NS have disturbed GH secretion but normal GH responsiveness, giving greater response on a similar dose of rhGH treatment compared to girls with TS.

The main finding of the present study was that NS-children showed higher response in IGFI as well as in growth, pointing to higher responsiveness to GH treatment compared to TS-girls.

In the NS-children but not in the TS-children these two GH effects showed great correlation.

The magnitude of mean growth response, independent of diagnosis and GH dose, was in parallel with findings in other patients with short stature without GH deficiency (32–34). Girls diagnosed with TS did not show as great response for IGFI as found in NS, whereas for growth the response was almost in parallel with similar doses of rhGH treatment, confirming findings by Romano et al. (16).

In our study, 60% of the variance in the first-year prepubertal growth response in NS could be explained by the chronological age at start, the GH-baseline level and the IGFI/IGFBP-3 ratio_SDS -the three selected pre-treatment variables by the multivariable regression models. In girls with TS, the rhGH dose and mean 24hGH became the selected variables explaining only 28% of the variance. The influence of rhGH dose in TS is in accordance with the KIGS outcome research database (1, 35).

NS boys showed several differences in the pattern of the 24hGH profile with higher peaks and trophs resulting in an overall higher 24hGH concentration when compared to healthy boys, whereas girls in the NS group had similar GH concentrations to normal girls. Despite the physiological positive feedback of the low IGF serum concentrations stimulating endogenous GH secretion, rhGH treatment resulted in normalized IGF concentration and prepubertal growth in both boys and girls with NS.

The approach of comparing NS to TS revealed that after adjustment for rhGH dose and GH-status, the strong correlation between changes in height_SDS with the changes in IGFs already after 10 days of rhGH treatment was significant in NS only. Indeed, the examination of short-term changes during the first weeks could be regarded as an IGFI generation test (36, 37). As shown in both non-GH deficient and GH deficient children the short-term changes in IGF, IGFBP3, and their ratio, were more informative for short and long-term growth response than the obtained IGFI_SDS-level per se on rhGH treatment (38, 39).

The lower increment in IGFI_SDS and IGFI/IGFBP3-ratio_SDS in TS could be interpreted as a substantial insensitivity for the IGF-response to GH in TS. This possibility was further supported by the fact that girls with TS, showed higher mean GH level, GH-baseline levels, amplitude, and number of GH peaks when measured with polyclonal antibodies (40). The abnormal proportion of non-22kDa GH isoforms (less active for growth) in TS than in normal growing children (41, 42) could be another explanation to a lower activation of the GH receptor in TS. Interestingly, here we found in four out of ten children with NS abnormally high proportion of non-22kDa isoform.

In both NS and TS, we found signs of an elevated GH secretion as measured with polyclonal antibodies, whereas the pulsatile pattern of GH with higher number of peaks was abnormal only in NS. These results are in accordance with findings by Noordam et al. (8) in 17 prepubertal NS children (13 boys and four girls) who also showed disturbances in the pattern of overnight GH secretion. They found low spontaneous GH max and high GH-baseline levels measured with overnight GH profiles using PULSAR (8), see review by Albertsson-Wikland and Rosberg (43). Half of their children had a mean overnight GH concentration below the lower limit of the normal range, whereas six instead had high GH baseline. Abnormalities in the GH secretion pattern with higher numbers of peaks and elevated GH-baseline could be due to a disturbance at the hypothalamic level such as an impaired balance between GH releasing hormone and somatostatin activity as implied from the present findings in Noonan boys. The fact that the mean GH level and the GH-baseline level are high could also indicate a partial resistance at GH post-receptor level and a reduced negative feedback mechanism.

A broad variation in individual growth response, independent of given rhGH dose, was found for both groups with NS and TS. This was previously found in children with other diagnoses as isolated GH deficiency, idiopathic short stature and small for gestational age (32–34). This is mainly caused by the individual responsiveness to GH, which can be mathematically estimated by prediction models for prepubertal growth response in children with GH deficiency, idiopathic short stature, small for gestational age (44, 45), and TS (1). For children with NS there is no specific model developed, leaving the clinician to use first-year growth response. We have previously shown that the first-year growth response incorporates the pre-treatment measures and can serve as a marker of individual responsiveness (3). In addition, the injection technique also influences the GH uptake (46). Daily deep subcutaneous injection is known to obtain high GH peaks and low/undetectable levels before next injection. Such a pattern has been shown to give an optimal signal for growth whereas continuous GH levels are more effective for IGF response (47).

In approximately 75% of children with NS, clinically diagnosed by phenotype, different gene mutations are detected (5). Most commonly, the PTPN11 gene mutation is found. In our group of 25 children with NS, we found somewhat higher proportions of identified PTPN11 versus other gene mutations and somewhat lower proportion of not identified genetic mutations. We subdivided the NS children into two groups - those with and those without the PTPN11 gene mutation – to explore the impact of the genetic basis on the GH/IGF axis and the GH responsiveness. The children with PTPN11 were shorter before treatment, thereby confirming precious results by Binder (48). However, no baseline IGFs or GH profile estimate differed according to the genetic background. On rhGH treatment, the two groups showed similar response for growth, whereas for IGFs those with PTPN11 showed a lower increment. Thus, we have not detected any impact of the various genetic bases in NS children for GH responsiveness for growth.

The study is not without limitations, such as the combination of data from several trials with somewhat different study design, although similar inclusion criteria. The different proportions of poly- and monoclonal antibodies would preferably been investigated in a majority of children for a stronger power and probably more significant results on growth response. A strength of the study was the relative high numbers of NS patients, besides 40 TS girls, that carefully were evaluated with a 24hGH profile and not only overnight sampling. Our one-centre study presents data on 12 males and 13 females with NS, which nearly reach the global previously published number of 30 NS patients being compared to TS (49). In addition, only two other papers with a total of 15 males and 6 females with NS have published spontaneous GH profiles overnight (8, 11). Another strength was that all participants remained prepubertal throughout the study period. This allowed accurate growth response evaluation without influence by pubertal growth, especially since it was estimated in relation to a prepubertal growth reference and not to a traditional total height reference with a mixed prepubertal/pubertal population for this age period (49).

This is the first study in NS to evaluate the 24hGH secretion, as the pattern, amount of total as well as the proportion of 22kDa and non-22kDa GH forms, the IGFI response and the growth response to rhGH. We found children with NS to have disturbances in the secretion pattern of GH, with better growth response to treatment with rhGH than girls with TS. In parallel, children with NS show a greater IGFI increase as a marker of high responsiveness. Despite the overall good growth response, the broad individual responsiveness to rhGH must be considered for a more tailored successful treatment. For this purpose, we suggest the first-year height gain to be used for personalized GH-dosing.

The datasets for this article are not publicly available because too few individuals could be a legal problem in identifying personal details. Requests to access the datasets should be directed to JD, jovanna.dahlgren@gu.se.

The studies involving human participants were reviewed and approved by Ethical committee at University of Gothenburg. Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin.

Both authors designed the study, developed the statistical analysis plan, and wrote the paper. KA-W is PI for all the studies used. All authors contributed to the article and approved the submitted version.

This work was supported by the Governmental grants under the ALF agreement (ALFGBG-812951, ALFGBG-719041) and an unrestricted grant from NovoNordisk in Denmark. The authors declare that this study received funding from NovoNordisk. The funder was not involved in the study design, collection, analysis, interpretation of data, the writing of this article or the decision to submit it for publication.

Free drugs were given from NovoNordisk for NS and Pharmacia/Pfizer for those TS girls who received the higher rhGHdose. None of the pharmaceutical companies influenced or had any impact on the data analyses nor the writing of this manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.