Treatment of short stature with recombinant human GH (rhGH) is well established, but response to and efficacy of the treatment varies from indication to indication and from patient to patient (1). Pharmacogenetics of the GH/IGF-1 axis attempts to explain some of the interindividual variability (2).

The growth hormone receptor (GHR) plays a key role in GH signaling and exists polymorphically in a full-length and an exon 3-deleted form (3). This is one of the most studied gene polymorphisms and has been presented in many association studies, occasionally yielding contradictory results (4, 5). In principle, in-vitro studies support better signal transduction by the exon 3-deleted receptor (4). Two metanalyses reported an advantage for carriers of one or two exon 3-deleted GHR alleles in terms of initial growth response to rhGH (6, 7). In addition, a variant of the IGFBP-3 promoter (-202 A/C) has been reported to be associated with serum IGFBP-3 level and response to rhGH (8, 9). Furthermore, a polymorphic (CA)n repeat 1kb upstream of IGF1 was reported to correlate with first year height velocity in children with severe GHD (10).

Recently, the methylation of the IGF1 promoter P2 has been linked to the response to rhGH treatment in children with idiopathic short stature (ISS) by a French study group (11). In principle, an increased amount of 5-methyl cytosine in the promoter region of a gene can cause partial or complete inactivation of gene expression (12). Methylation of CpG islands in the promoter region of IGF1 could therefore affect IGF-1 production and thus responsiveness to GH (13). Ouni et al. reported that the methylation of specific cytosines in the IGF1 P2 promoter in blood leucocytes negatively correlates with the transcriptional activity of IGF1 and may explain approximately 25% of variability in responsiveness to rhGH in children with idiopathic short stature (11, 14, 15). These results have not been confirmed so far.

The aim of this project was to analyze the clinical relevance of IGF1 promoter P2 methylation at DNA positions -137 and +97 in a cohort of children with GHD, Turner syndrome and SGA short stature treated with rhGH. In addition, we investigated the stability of the specific DNA methylation sites in blood of the patients longitudinally.

A total of 92 patients with short stature and rhGH treatment donated residual blood from routine examinations for research in the Pediatric Endocrinology section of the University Children`s Hospital Tübingen since 2015. They were diagnosed with GHD (n=40), short stature after being born small for gestational age (SGA short stature) (n=36), or Turner syndrome (n=16). Height was measured using a wall-mounted stadiometer system Dr. Keller II (Längenmeßtechnik GmbH Limbach, Germany). Growth hormone deficiency was diagnosed in children with growth retardation, delayed bone age, pathologically low IGF-1 and two pathological GH stimulation tests. SGA short stature was defined by birth weight or birth length < -2 SDS and height < -2.5 SDS at 4 years of age after exclusion of other specific diagnoses. Turner syndrome was diagnosed in short girls with a karyotype that partially or completely lacked one sex chromosome. Bone age was determined using the atlas by Greulich and Pyle (16). For correlation with promoter P2 methylation, height velocities were excluded from children who were < 4 years of age at the start of rhGH treatment.

Both, the establishment of the biobank (residual blood) and this specific study were independently approved by the Ethics Committee of the Medical Faculty of the University of Tübingen (259/2014BO1 and 050/2020BO2). The study was conducted in accordance with the Declaration of Helsinki of the World Medical Association. All patients´ caregivers provided written informed consent.

The characteristics of the patients and their response to treatment are shown in Table 1 . Patients were relatively young at treatment start for all three diagnoses. The response to rhGH treatment was above average in terms of delta height velocity and studentized residuals. A total of 443 patient samples could be used for genetic and epigenetic analysis.

The methylation of the P2 promotor of IGF1 at positions -137 and +97 could be determined in the blood of 40 children with GHD, 36 children with SGA short stature, and 16 children with Turner syndrome. The mean (SD) values at position -137 were 59.5% (6.6) in GHD, 57.6% (7.2) in SGA short stature, and 64.1% (8.1) in Turner syndrome. The percentages at position +97 were as follows: 17.8% (4.8) in GHD, 18.3% (5.4) in SGA short stature, and 22.6% (4.3) in Turner syndrome.

The haplotypes of the GHR polymorphism were as follows: in GHD wt-wt (60%), wt-del3 (40%) and del3-del3 (0%), in SGA short stature wt-wt (44.5%), wt-del3 (44.5%) and del3-del3 (11%), in Turner syndrome wt-wt (38%), wt-del3 (56%) and del3-del3 (6%). In total, 50% of patients were wt-wt, 45% were wt-del3 and 5% were del3-del3.

Longitudinal analysis of the methylation over a period of 0.5 years before treatment to 12.0 years during treatment with rhGH showed a variance of 7.1% (-137) and 4.9% (+97), but no specific trend toward lower or higher values over time in any of the three diagnostic groups. There was no unidirectional change in samples from 0, 6 and 12 months during the first year of treatment. There was no longitudinal trend in methylation of the P2 promoter of IGF1 at position -137 ( Figure 1 ). Importantly, the inter-assay variability of the methylation of our control samples was 9.5% (-137) and 6.6% (+97).

The methylation of the P2 promoter of IGF1 at position -137 was not correlated with delta height velocity in any of the three diagnostic groups ( Figure 2 ). The same result was found for the +97 position (data not shown).

The methylation of the P2 promoter of IGF1 at position -137 showed no relationship with the studentized residuals of the three diagnostic groups ( Figure 3 ). The same result was found for the +97 position (data not shown).

The IGF-1 concentrations at baseline and their increase from baseline to 12 months of treatment (delta SDS) also did not correlate to the P2 promoter methylation ( Figures 4 and 5 ).

Multiple linear regression analysis including the GHR genotype (exon 3 deletion versus no exon 3 deletion) did not show any relationship between outcome parameters and the examined genetic characteristics (data not shown.). In addition, the GHR genotype did not correlate to IGF-1 serum concentration or GH responsiveness (data not shown).

Methylation of the IGF1 promoter P2 at two specific CG-sites was measured in the blood of children with GHD, SGA short stature and Turner syndrome treated with rhGH. Promoter methylation did not correlate with first-year response in terms of height velocity increment or studentized residuals based on a prediction model at either position -137 or position +97. Furthermore, baseline IGF-1 serum concentrations and increases during rhGH treatment did not correlate with IGF1 P2 promoter methylation. Our analysis is in contradiction with previous reports from Ouni and colleagues that reported such correlations observed in children with low-normal stature or ISS treated with very high doses of rhGH (11).

Ouni et al. used the same source, peripheral blood leucocytes, for epigenetic analysis. In contrast to this study, blood was collected only before the start of rhGH treatment by the French group, whereas 94% of our samples were collected during rhGH treatment. However, the intraindividual changes of promoter methylation observed in our study were small after the start of rhGH treatment and showed no trend. The pyrosequencing protocols were identical. The French group observed a mean of 49% methylation at the -137 site of promoter P2, and 17% methylation at the +97 site. Here, the methylation values were 60% at -137 and 19% at +97, approximately 20% higher.

Ouni et al. treated 136 children with ISS (or low-normal stature) with a high pharmacological rhGH dose of 67 µg/kg*day and observed an inverse relationship between first-year increment of height velocity and the methylation at the IGF1 P2 methylation site -137 in blood cells (11). They calculated that 25% of the variability of the height velocity increment could be explained by the methylation at this IGF1 promoter site. Our study cohort did not include children with ISS, but GHD, SGA short stature and Turner syndrome. The height velocity increments were measured in the same way. Our patients were by 2.5 years younger and the sample size was smaller than in the French studies. The rhGH doses here ranged from 20 to 60 µg/kg*day, which was lower than in the French study. Although the studied children and the treatment dose were different, one would expect at least a trend in the same direction if the reported correlation is not specific for ISS and for very high doses of rhGH. Moreover, our analysis of responsiveness including relevant outcome parameters of individual patients (studentized residuals of prediction) showed no lower response to rhGH in patients with a higher methylation of IGF1 P2 promoter sites.

Similarly, Ouni et al. reported an inverse correlation of basal serum IGF-1 concentration with the methylation at -137 of the IGF1 P2 promoter in a mixed group of children with normal or idiopathic short stature (14). Moreover, the short-term IGF-1 increase in the IGF-1 generation test (SDS and concentration) was shown to be higher in children with lower methylation of the IGF1 promoter (11, 15). We analyzed baseline IGF-1 and the increment after 12 months of rhGH treatment and found no correlation with the methylation status of the IGF1 promoter. In conclusion, there are significant differences in study conditions between the French studies and our study, including different rhGH dose, different diagnoses, and different timing of IGF-1 observation, which may explain the conflicting results.

In a second part of the study, we examined the stability of the methylation marks of the IGF1 P2 promoter in blood cells over time during rhGH treatment. The data suggest good stability and no specific change over time. However, we observed intraindividual variation in long-term results, which may possibly reflect changes in the number of different leucocyte populations in blood specimens.

In conclusion, we could not confirm that IGF1 P2 promoter methylation is an important epigenetic locus for GH responsiveness in children with GHD, SGA short stature and Turner syndrome. This nonconfirmatory study will hopefully motivate other research groups to publish their findings or to perform new analyses, because more data are needed for a definitive judgement in this field of pharmacogenetics.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

The studies involving human participants were reviewed and approved by Ethics Committe of the Medical Faculty at the University of Tübingen. Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin.

GBi and AA contributed to conception and design of the study. RS, AA and DII organized the database. AA, DII, CU, KW, SP and VN performed or supervised the (epi-)genetic analyses. AA and GBl performed the statistical analysis. AA and GBi wrote the first draft of the manuscript. All authors contributed to manuscript revision, read, and approved the submitted version.

The research project was supported by a grant from Novo Nordisk Pharma GmbH, Mainz, Germany (Investigator Sponsored Study U1111-1248-5097). 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. All authors declare no other competing interests. DII was supported by DAAD funding program “Research Stays for University Academics and Scientists” (Number 57552334, year 2021).

Technical support of pyrosequencing by Mara Thomas, Nora Knoblich and Ariane Wiegand was greatly appreciated. We are thankful to Joachim Woelfle for fruitful discussion during designing the study.

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.