Effects of recombinant human growth hormone (rhGH) administration on body composition and cardiovascular risk factors in obese adolescent girls

Participants were recruited at Massachusetts General Hospital (MGH) between September 2010 and October 2012 through area pediatric and obesity clinics. Of 32 girls and young women 13–21 years old assessed for study eligibility, 22 obese adolescents met inclusion criteria and were randomized. Inclusion criteria comprised (i) a bone age ≥14 years, (ii) body mass index (BMI) > 95th percentile for age (based on the 2000 Centers for Disease Control and Prevention Growth Charts) [15], or >30 kg/m² if age >18 years, (iii) insulin like growth factor-1 (IGF-1) level below the median for pubertal stage or age, (iv) abdominal obesity with a waist to hip ratio (W/H) >0.85, and (v) stable weight (<5 kg change in weight in the prior 3 months). Exclusion criteria included diabetes mellitus, untreated thyroid dysfunction, chronic renal insufficiency, current or past malignancy, syndromic obesity, pregnancy, breast-feeding, and use of medications known to alter glucose metabolism or body composition (contraceptive pills, daily glucocorticoid use, metformin, sibutramine and Orlistat). Additionally, we excluded girls with new (<6 months) or unstable dosing (dosage change within 3-months) of antipsychotic medications known to cause weight gain. Baseline characteristics for a subset of study participants have been previously published [16].

Eleven subjects were randomized to active rhGH treatment {rhGH (+)} and 11 to rhGH negative {rhGH (-)} treatment. Sixteen subjects {10 rhGH (+) and 6 rhGH (-)} completed the 3-month visit, and 12 subjects {5 rhGH (+) and 7 rhGH (-)} finished the 6-month study period (one rhGH (-) subject missed the 3-month visit but completed other visits). No subject dropped out because of side effects. Of subjects who withdrew (n = 10), one chose to begin an oral contraceptive, 3 withdrew because of personal obligations, and the remaining were considered lost to follow-up (n = 6). Study participants who did or did not complete the study did not differ for baseline characteristics.

The study was a 6-month, single-blind, randomized trial, conducted at the Clinical Research Center of MGH. The screening visit included a history and physical examination and blood sampling to ensure participants met inclusion criteria and to rule out exclusion criteria. Diabetes was ruled out via an oral glucose tolerance test (OGTT). Eligible subjects returned for the baseline visit, visits at 1 and 2 weeks (for dose adjustment), and at 1, 3 and 6 months. Subjects were randomized 1:1 to receive once daily subcutaneous rhGH or placebo injections (Somatropin, Genentech, Inc., San Francisco, CA, USA), and taught to self inject medication daily for 6-months. Due to non-availability of placebo (and inability for new placebo to be manufactured) after June 2012, subjects subsequently randomized to placebo (n = 4) were instead randomized to no treatment. Dieticians and the radiologist who performed study related procedures as well as all subjects who enrolled prior to June 2012 (N = 15) remained blinded to randomization status for the entirety of the study. After the placebo expiration necessitated the change in the study design, one subject randomized to rhGH (+) subject and one subject randomized to rhGH (-) treatment dropped out. The subject randomized to rhGH (-) treatment withdrew due to the initiation of an oral contraceptive while the subject randomized to rhGH (+) was lost to follow up after the 3 month visit.

The starting rhGH dose was 0.4 mg and increased at week-1 and week-2 to 0.6 mg and 0.8 mg respectively. The dosing was based on a dose of 12.5 mcg/kg/day for a 16-year-old girl (mean for bone ages 14–18 years) weighing 67 kg (85th percentile for age) [17]. This dosage (0.8 mg at week-2) is at the lower end of the 12.5–25 mcg/kg/day dose used in studies of GH deficient adolescents transitioning from pediatric (40–45 mcg/kg/day) to adult (2–6 mcg/kg/day) replacement rhGH doses [17, 18]. We opted for this lower dose as we expected lower GH requirements in this population that is relatively, rather than completely, GH insufficient. The dose was adjusted as needed by 20% at the 1 and 3-month visits to achieve IGF-1 levels in the upper half of the normal range for pubertal stage. When target IGF-1 was achieved, the individualized dose was continued.

Height was measured as the average of three measurements to the nearest 0.1 cm on a single calibrated wall-mounted stadiometer. Participants, wearing a hospital gown, were weighed to the nearest 0.1 kg on a single calibrated electronic scale. BMI was calculated as weight (in kg) divided by height (in meters²) and BMI standard deviation scores (SDS) determined from 2000 CDC charts [15]. Waist measurements were taken with a plastic tape measure to the nearest 0.1 cm at the level of the iliac crest and umbilicus; the maximum hip circumference was measured. All measurements were taken at the end of expiration with the subject standing. Waist-to-hip ratio (W/H) was calculated as the iliac divided by the hip circumference measurement. An ophthalmoscope was used to rule out papilledema at study onset, and at the 3 and 6-month visits, or if any subject complained of headaches, vision changes, nausea or vomiting. Subjects had a left hand x-ray to assess bone age [19]. Pubertal stage was determined according to the criteria of Tanner [20].

Magnetic resonance imaging (MRI) at the level of L4 was used to determine visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT) at baseline and 6 months; total abdominal adipose tissue (TAT) was calculated as VAT + SAT [21]. MRI data are available for 11 completers (due to scheduling conflicts one no-treatment subject was unable to perform the MRI component). Body composition was also obtained by dual energy x-ray absorptiometry scans (DXA) at the baseline, 3 and 6-month visits. DXA (Hologic QDR-Discovery A; Hologic Inc., Waltham, MA software version APEX 3.3) was used to assess percent body fat, total, trunk and extremity fat and total and extremity lean mass.

Subjects met with nutritionists of the Clinical Research Center at the baseline, 3 and 6-month visits for (i) basic lifestyle counseling including healthy eating habits and optimizing exercise, and (ii) assessment of prior activity using the Modified Activity Questionnaire (MAQ) [22]. To avoid confounding variables, further dietary or exercise restrictions were not imposed.

Fasting blood samples were drawn for IGF-1 at each visit for rhGH dose adjustment, and a 75-g, 2-hour OGTT performed at screen, 3 and 6-months to determine whether rhGH administration had deleterious effects on glucose tolerance. Fasting blood samples were also assessed for cardiovascular risk markers [lipids, high sensitivity C-reactive protein (hs-CRP), soluble intercellular adhesion molecule-1 (sICAM-1)], HbA1C, glucose and insulin levels at baseline and 6-months.

JMP Software (v10: SAS Institute, Inc., Cary, NC) was used for analysis. Results are reported as means ± SD. Baseline and mean 6-month change were compared using the Student t-test or Wilcoxon rank sum test depending on data distribution. A paired t-test was also used to compare baseline and 6 month values within groups. Parametric (Pearson) or nonparametric (Spearman) correlations were used as appropriate to determine associations between variables that were or were not normally distributed. Significance was defined as a two-tailed p-value of <0.05. Multivariate regression models were constructed to determine whether significances for changes in endpoints persisted after controlling for weight changes or baseline values of the endpoint (i.e. HbA1c).

Baseline characteristics are summarized in Table 1. Study participants had a mean age of 16.6 ± 2.4 years, mean weight of 100.1 ± 20.5 kg, and a mean BMI of 38.5 ± 7.5 kg/m². Treatment groups did not differ for baseline characteristics except for a slightly lower HbA1C in the rhGH (-) group. There were no significant baseline differences noted between the blinded (N = 15) and unblinded (N = 7) groups with one exception; the W/H ratio was higher in the group who remained blinded to their treatment allocation (0.96 ± 0.06 vs. 0.90 ± 0.07, p = 0.05). All subjects were Tanner Stage V for breast development at the baseline visit.

No differences were noted between rhGH(+) and rhGH(-) groups for changes in calories consumed (142.6 ± 961.0 vs. 193.5 ± 644.0 kcal, p = 0.93), percentage of calories from fat (-8.09 ± 12.04 vs. 0.09 ± 12.95%, p = 0.36), or activity levels (1.64 ± 6.4 vs. 8.83 ± 9.48 hours/week, p = 0.18).

Changes in IGF-1 trended higher in the treatment group (Table 2). At the 6-month visit, mean IGF-1 in the rhGH(+) group was non-significantly higher than in the rhGH(-) group (234 ± 24.3 vs. 185.6 ± 20.5 ng/mL, p = 0.16).

Body composition changes across treatment groups are presented in Table 2. Although changes in waist circumference and W/H ratio did not differ across groups over the study duration, the increase in IGF-1 between 3–6 months correlated with the decrease in W/H ratio over 6-months (r = -0.74 p = 0.009). A similar finding was also seen in the negative correlation between change in IGF-1 over 3-months and change in body fat percentage over 3-months (r = -0.65 p = 0.009) (Figure 1). SAT and TAT decreased non-significantly in the treatment group compared to an increase in the control group, and VAT showed a negligible increase compared to the control group.

At 6-months, subjects in the active group had significant reductions in HbA1c compared to controls (Table 2). This reduction trended to remain significant after controlling for baseline HbA1C (p = 0.06). Subjects in the active group also trended to have greater reductions in 2-hour glucose levels on the OGTT (Table 2). No difference was noted between the groups for 6-month changes in insulin levels. Thus, we did not see the anticipated deleterious effects of rhGH on glucose tolerance.

Adverse events are presented in Table 3. There were no serious adverse events related to study participation. rhGH was well tolerated and no subject required a dose reduction. No subject who complained of headaches, nausea or vomiting had a history or physical exam findings suggestive of pseudotumor cerebri.