Thirty consecutive patients with PWS (14 men, 16 women, aged 18–28 years), referred to the Department of Endocrinology Medical University of Warsaw in the years 2016-2021, were included in the study. All patients had been previously treated in the Departments of Pediatric Endocrinology and Diabetes. The diagnosis of PWS was based on the clinical phenotype and genetic testing.

None of the patients had been treated with systemic or topical glucocorticosteroids (GCS) for the last six months. Twenty-two patients (73.3%; 12 women and 10 men) had been treated with recombinant human growth hormone (rhGH) in the last 12 months. GH therapy was continued in all previously qualified subjects. The IGF-1 levels were age- and gender-appropriate at baseline and follow-up. Due to hypogonadotropic hypogonadism, all patients received hormone replacement therapy (estradiol and progesterone in females or testosterone in males). Moreover, 16/30 (53.3%) patients received LT-4 supplementation. Dyslipidemia was identified in 14 out of 30 patients (46.7%) and was managed with dietary interventions; no statins were prescribed. All patients had normal vitamin D levels while on supplementation. None of the patients had anemia or iron deficiency.

Physical examination included measuring height, weight and body fat percentage in fasting conditions. Standing height was determined by a stadiometer WPL150. Body weight (to nearest 0.1kg), and body fat percentage were measured using the electrical bioimpedance method with Tanita Body Composition Analyser Type TBF- 300 MA (17, 18). Body Mass Index (BMI) was defined as weight (kg)/height (m²). Following the World Health Organization criteria, the BMI cut off points of 18.5-25.0 to define normal weight, 25.0-30.0 to define overweight and >30.0 to define obesity, were used. Physical examination also included determination of arterial pressure (including orthostatic hypotension) and heart rate.

Any concomitant diseases and additional treatments were recorded ( Table 1 ).

The study protocol was approved by the Medical University of Warsaw Bioethics Committee (AKBE/88/2020). Written informed consent was obtained from the patients and their caregivers. The study was registered on the ClinicalTrials.gov (Identifier Number: NCT04700644).

Adrenal function was assessed using the high-dose short Synacthen test (HDSST) after the transition from pediatric to adult care (3) (VISIT I). Clinical assessments were then performed at the 6-month and 12-month follow-up visits (VISIT II and III, respectively).

At VISIT I, the adrenal function was assessed by:

HPA axis impairment (HPAI) was suspected in patients:

In cases of suspected HPAI, all therapeutic options (close monitoring, hydrocortisone (HCT) stress-dose and low dose HCT treatment) were thoroughly discussed with patients and their caregivers. Hydrocortisone supplementation was initiated in dose of 15mg/m², resulting in 10mg/day in two divided doses in the HPAI group (19, 20). During the follow-up, patients were asked to report possible side-effects of HCT therapy, including: infections, weight gain, increase in the amount of adipose tissue and metabolic abnormalities (impaired fasting glucose and lipid disorders).

None of the patients were on GLP-1 receptor agonists at the time of the study, although some were taking metformin.

All blood samples were immediately delivered to the laboratory and assayed. Blood samples for ACTH measurement were placed on ice until assayed.

All hormone assays were analyzed in the Central Laboratory of the Central Clinical Hospital of the University Clinical Centre at the Medical University of Warsaw. Cortisol levels were measured by electrochemiluminescence ECLIA method using Elecsys Cortisol II Cobas^® diagnostic kits (Roche Diagnostics Mannheim, Germany). The immunoassay for the in vitro quantitative determination of cortisol in human serum or plasma has a measuring range of 1.5–1750 nmol/L (0.054-63.4 μg/dL). Reference values for morning cortisol were: 133–537 nmol/L (4.82-19.5 μg/dL). DHEA-S concentration was measured by the electrochemiluminescence ECLIA method using Elecsys DHEA-S Cobas^® diagnostic kits (Roche Diagnostics Mannheim, Germany). The assay for the in vitro quantitative determination of DHEA-S in human serum or plasma has a measuring range of 0.003-27 μmol/L (0.010–100 ng/mL). Reference values were 5.72–13.35 μmol/L (211-492 μg/dL). ACTH was determined by electrochemiluminescence ECLIA method using Elecsys ACTH Cobas^® diagnostic kits (Roche Diagnostics Mannheim, Germany). Reference range for baseline ACTH was 1.6–13.9 pmol/L (7.22–63.3 pg/mL).

Values are presented as median with ranges for continuous variables, and as a number of patients and percentage of sample for categorical variables. To assess the associations between variables the Mann–Whitney U test was used for continuous variables. The chi-square and the Fisher exact tests were used for categorical variables. Correlation was tested with Spearman’s method. The level of significance was set at p<0.05. The dependent groups were compared using Wilcoxon mathed–pairs test for continuous variables. Statistical analyses were performed by STATISTICA software ver.13.1 (TIBCO Software, Palo Alto, California, United States of America).

There were no significant differences between the studied groups regarding anthropometric parameters (age, sex, weight, body mass index), genetic abnormalities, concomitant diseases and additional treatments – Table 1 .

The median (range) morning cortisol, ACTH and DHEA-S levels of the HPAI group were  173.9 nmol/L (91.0-359), 4.71 pmol/L (2.44-11.4) and 8.16 μmol/L (3.64-12.9), respectively. Cortisol levels below 133 nmol/L were found in 6 (20.0%) of cases. None of the patients had ACTH levels below 1.60 pmol/L. There were significant differences in morning (baseline) cortisol values between the HPAI and the Control group [median (range), 139 nmol/L (91.0-177), 238 nmol/L (166-359), respectively)] (P<0.001 ( Figure 1A ). There were no significant differences in the ACTH levels between the HPAI and the Control group [median (range), 4.40 pmol/L (2.44-6.38) and 5.08 pmol/L (2.46-11.4), respectively)] (P=0.06) ( Figure 1B ). Basal DHEA-S also did not differ between the two groups [median (range), 7,9 umol/L (5,2-10,4) and 8,3 (3,8-12,9), respectively)] (P=0.37) ( Figure 1C ). There were no significant differences in hormonal levels between the patients with UPD and patients with 15q (11-13) deletion and between two sexes.

Results of the HDSST are shown in Figure 2 . Fourteen of the 30 patients (46.7%) showed a 30-min cortisol peak <500 nmol/L, and were assigned to the HPAI group. Five of these patients had a 30-min cortisol response below 440 nmol/L and 6 had basal cortisol levels below 133 nmol/L. Peak cortisol levels at 30’ were significantly lower in the HPAI compared to the Control group, the median (range) being: 452 nmol/L (381-532) and 607 nmol/L (555-731), respectively (P<0.001). All patients but one (in the HPAI group) showed a 60-min cortisol peak to HDSST ≥500 nmol/L. However, peak cortisol levels at 60’ were also significantly lower in the HPAI group when compared with the Control group, the median (range) being: 548 nmol/L (486-654) and 694 nmol/L (635-822), respectively (P<0.001). Also Δ cortisol 30’ was significantly lower in the HPAI compared to the Control group. Median Δ cortisol 30’ was 317 nmol/L and 371 nmol/L, respectively (P=0.002). But there was no significant difference regarding Δ cortisol 60’ in the HPAI and the Control group with median 410 nmol/L and 445 nmol/L, respectively (P=0.07). Correlation analysis revealed that basal cortisol was positively correlated with cortisol levels at both 30’ and 60’ of the HDSST (r = 0.872, P < 0.001 and r = 0.829, P < 0.001, respectively).

Signs and symptoms suggestive of hypothalamic-pituitary-adrenal axis impairment in the study group at I VISIT are presented in Table 2 . Patients in the HPAI group reported muscle weakness and myalgia significantly more often than the Control group ( Table 2 ). None of the patients had been previously hospitalized due to an adrenal crisis or other life-threatening conditions.

All patients with HPA axis impairment agreed to receive HCT supplementation resulting in a daily dose of 10 mg/day, taken in two divided doses. After 6 and 12 months of HCT therapy patients in the HPAI group improved significantly in terms of fatigue, myalgia and muscle weakness, which was confirmed by the observation of their caregivers ( Table 3 ). After 12 months all but one patient decided to continue HCT supplementation at a lower dose (5 mg per day). No significant adverse effects of HCT treatment were observed within the 6 and 12 months of observation. Body weight and BMI measurements evaluated at 6 and 12 months of treatment showed no significant differences from baseline. Moreover, the percentage of body fat decreased significantly after 6, and 12 months of the treatment ( Table 3 ). A similar trend in weight and fat tissue was observed in both rhGH treated and not treated patients.

Similar to the HPAI group, after 6 and 12 months of observation patients in the Control group improved in terms of fatigue, myalgia and muscle weakness ( Table 3 ). Body weight and BMI measurements evaluated at 6 and 12 months of observation showed no significant differences from baseline. The percentage of body fat did not differ significantly after 6 and 12 months of the observation ( Table 3 ). A decrease in body fat percentage was observed in some subjects, possibly due to the increase in muscle strength and physical activity (reported by caregivers). Increased daily physical activity was noted in all participants.

Before starting the HCT treatment, 7 patients with HPAI had IFG treated with metformin, which did not get worse after 6, and 12 months of the HCT treatment. None of the other patients developed new carbohydrate disorders during 12 months of the HCT treatment.