Introduction

Phelan–McDermid syndrome (PMS) is caused by deletions in the long arm of chromosome 22 which include the SHANK3 gene (MIM: 606230), or by pathogenic sequence variants in SHANK3 [1,2,3,4]. PMS is associated with developmental delays, intellectual disability, and autism spectrum disorder (ASD), in addition to renal, cardiac, and gastrointestinal abnormalities, hypotonia, and dysmorphic features [5]. SHANK3 has been established as the critical gene in PMS [1,2,3,4, 6] and appears to account for ~ 0.5% of ASD [7]. SHANK3 encodes for a master scaffolding protein in the post-synaptic density of excitatory synapses and is responsible for the formation and maintenance of synapses [8]. As such, SHANK3 and associated pathways represent important targets for intervention.

Evidence from both preclinical and clinical studies suggests that insulin-like growth factor-1 (IGF-1) can reverse deficits in synaptic plasticity and motor learning in mouse and human neuronal models of PMS [9, 10]. A clinical trial with IGF-1 in children with PMS also showed improvement in social withdrawal and restricted behaviors, both core features of ASD [11]. Additional evidence of the utility of IGF-1 comes from animal, human, and human neuronal studies of Rett syndrome, another rare genetic disorder associated with ASD, where IGF-1 was effective in reversing phenotypic features [12,13,14,15,16].

IGF-1 is released mainly by the liver upon growth hormone stimulation and enters the brain from the circulation to promote brain vessel growth [17], neurogenesis, and synaptogenesis [18]. Once IGF-1 binds to the IGF-1 receptor, activation of the PI3K/mTOR/AKT1 and MAPK/ERK pathways induces its downstream effects [19]. Treatment with IGF-1 is generally administered twice daily via subcutaneous injection and requires careful monitoring due to numerous risks, including hypoglycemia. Further, IGF-1 is challenging to manufacture and while commercially approved for short stature due to primary IGF-1 deficiency, it is costly and not readily available. However, IGF-1 levels can be increased intrinsically by growth hormone [20] without the risk of hypoglycemia. Recombinant human growth hormone (rhGH) has an excellent safety profile and approved indications in pediatric and adult populations. One recent case report also supports the use of rhGH in PMS [21]. For these reasons, rhGH was chosen for this trial with the primary aims of demonstrating the feasibility of increasing IGF-1 levels in the blood and establishing safety in PMS. Furthermore, we sought to explore signals of efficacy using a battery of clinical outcome assessments, including the Aberrant Behavior Checklist—Social Withdrawal subscale (ABC-SW) [22] as the primary clinical outcome. The ABC-SW subscale was chosen based on results from the previous clinical trial with IGF-1 in PMS [11].

Methods

This study was approved by the Program for the Protection of Human Subjects at the Icahn School of Medicine at Mount Sinai, and all caregivers provided written informed consent.

Inclusion/exclusion criteria

Participants were required to have a confirmed genetic diagnosis of PMS and be between 2 and 12 years of age. Participants were excluded if they had closed epiphyses, active or suspected neoplasia, intracranial hypertension, hepatic insufficiency, renal insufficiency, cardiomegaly/valvulopathy, or allergy to growth hormone or any component of the formulation.

Drug administration

rhGH was administered in its commercially available form as somatropin (Zomacton). Caregivers were trained by a pediatric endocrinologist (Sethuram, S) to administer rhGH subcutaneously, through demonstration and written material. rhGH was given once daily for 12 weeks using an open-label design. Doses were based on standard clinical practice for children who are not growth hormone deficient with a target dose of 0.3 mg/kg/week. All participants were initiated on half the target dose (0.14–0.16 mg/kg/week) for two weeks as a safety precaution and then increased to a full dose for the remaining 10 weeks. IGF-1 levels were measured every 4 weeks, and IGF-1 Z scores were used to guide titration of rhGH dose using two standard deviations (SD) above the population mean as the target.

Safety measures and laboratories

Medical and psychiatric history was collected prior to the onset of the trial. Safety laboratories, physical examinations, and IGF-1 values were collected at the baseline visit and at each follow-up visit: weeks 4, 8, and 12. Adverse events were collected at every visit using the Systematic Longitudinal Adverse Events Scale (SLAES).

Clinical measures

The primary clinical outcome was the ABC-SW subscale (ABC-SW) [22]. Additional clinical outcome assessments were used to capture a range of ASD-related symptoms, including the Repetitive Behavior Scales—Revised (RBS-R) [23], the Sensory Profile (SP) [24], other ABC subscales (Table 2), and the Clinical Global Impression—Improvement scale (CGI-I) [25].

Statistical analyses

Nonparametric Wilcoxon signed-rank tests were used to evaluate differences in clinical outcomes between baseline and week 12. All tests of statistical hypotheses were done on the two-sided 5% level of significance. We selected a single primary efficacy variable (ABC-SW) a priori and did not adjust for multiplicity of statistical tests. All raw p values are presented to allow an adjustment post hoc (Table 3). In the case of missing data, we used the last observation carried forward. The sample size was not based on statistical criteria and was determined by feasiblity for this pilot study.

Results

Participants

This trial was conducted from September 2019 to June 2020 and terminated early due to COVID-19; the original recruitment target was 10 participants. Six participants were screened, and all met inclusion criteria and were enrolled. Participants (2 males; 4 females) were between 3.2 and 11.4 years of age (7.5 ± 3.2). All participants except one female were pre-pubertal. The one child who was pubertal on physical and biochemical evaluation did not reach menarche. At baseline, all children were of average weight (− 0.85 ± 1.15 SD), height (− 1.38 ± 0.75 SD), and body mass index (-0.82 ± 1.27). All bone ages were within the normal range. Baseline IGF-1 Z scores varied between − 1.2 and 2.3 (Table 1).

Table 1 rhGH dose in mg/kg/week and IGF-1 Z scores
Full size table

Safety

Recombinant human growth hormone was generally well tolerated, and there were no serious adverse events (Table 2). On average, participants experienced approximately five treatment emergent adverse events. One participant experienced gait changes, and rhGH was terminated early at week 11 out of an abundance of caution due to the risk of slipped capital femoral epiphysis. The participant was evaluated by their pediatrician, and no additional workup was deemed necessary; gait normalized within 2 days after stopping rhGH and without further sequelae. Another participant required dose reduction due to crying spells. Crying spells in all three participants were attributed to increased emotional lability. There were no clinically significant abnormalities on laboratory blood work.

Table 2 Adverse events
Full size table

Efficacy

There was an improvement in our primary clinical outcome, the ABC-SW subscale, between baseline and week 12 (p = 0.028) (Fig. 1). There was also an improvement in hyperactivity using the ABC hyperactivity subscale (p = 0.027), and in overall sensory symptoms as measured by the short sensory profile total score (p = 0.042). Overall, there was global improvement as measured by the CGI-I (p = 0.023). There were no significant changes in other clinical domains (Table 3).

Fig. 1
figure 1

Domains of clinical improvement. Lower ABC scores indicate improved behavior, and higher SSP scores indicate improved behavior

Full size image
Table 3 Summary statistics for clinical outcomes
Full size table

Discussion

The results of this pilot open-label clinical trial demonstrate that standard clinical doses of rhGH increased levels of IGF-1 in children with PMS by at least 2SD from baseline for all participants; final levels of greater than or equal to 2SD were reached in all except one participant. Further, we show that rhGH was well tolerated without serious adverse events. As rhGH is already FDA-approved and established as safe in children with growth-related problems and in adults with growth hormone deficiency, these results provide preliminary evidence of safety in a new patient population without specific growth issues. rhGH treatment was also associated with clinical improvement that parallels the effects of IGF-1 on social withdrawal in this population. In addition, rhGH was associated with benefits in hyperactivity and sensory symptoms, all leading to global improvement based on the CGI-I. Studies of rhGH in PMS are ongoing using a randomized, placebo-controlled, crossover design. In addition, it will be critical to discover biomarkers to predict treatment response to rhGH in PMS, and potentially, within subgroups of ASD more broadly.

Limitations

Results should be interpreted with caution given the small sample size and open-label design of the study.

Conclusions

Taken together, these findings support the development of rhGH as treatment for children with PMS. Future studies of the effects of rhGH in PMS using an adequately powered placebo-controlled design are warranted.