GeNeSIS was an open-label, multinational, prospective observational research programme the purpose of which was to assess the long-term safety and effectiveness of GH therapy (Humatrope®, Eli Lilly and Company, Indianapolis, USA) in children with short stature. Patients were diagnosed and managed according to standard paediatric endocrine practice. The programme was conducted in accordance with the ethical guidelines of the Declaration of Helsinki, received institutional board approval, and met all applicable regulatory requirements in the participating countries. The parents or legal guardians of each patient provided written consent for data collection, processing and publication prior to enrolment.

The study included patients undergoing or that were going to start treatment with GH for growth promotion. Patients with a previous history of neoplasm or a SHOX deficiency disorder were only included if they had never received GH therapy before. Patients with closed epiphyses were not eligible for entry, although those whose epiphyses closed during the study were allowed to continue participating in the study. In Spain, 56 centres with paediatric endocrinology units participated in the study.

Our study analysed the data for Spanish prepubertal and pubertal children included in GeNeSIS from its beginning in 1999 to September 2012. All decisions involved in the diagnosis and management of the patients were made by their attending physicians, and the data are presented as documented by the investigators. We assessed the auxological parameters of a subset of this cohort of Spanish children diagnosed with GHD that had not received treatment with GH prior to enrolling in the study and for whom data for four years’ followup was available. We also evaluated the final height (when patients were nearing adulthood) of another subset treated with GH during the study, independent of whether or not they had received GH therapy prior to enrolment.

The following variables were assessed at the time of enrolment: chronological age, target height (mean height of father and mother±6.5 depending on the patient's sex), weight, body mass index, bone age, and peak GH value in the GH stimulation tests. The initial GH dose for each patient was documented. At enrolment time and also once a year during followup, the patient's height, growth velocity (GV), bone age and pubertal status (Tanner stage) were assessed. For patients that reached their final heights, the final height and the difference between final height and target height were recorded. We calculated standard deviation scores (SDS) based on the 2000 US National Center for Health Statistics standards.20

We assessed safety based on the adverse events and serious adverse events (SAEs) experienced by the Spanish patients treated with GH reported by the investigators. We defined serious adverse events as those resulting in death, hospitalisation, permanent or significant disability, that were life threatening or otherwise considered significant by the investigator. We classified adverse events according to the definitions of the Medical Dictionary for Regulatory Activities (MedDRA) version 11.0. The researcher determined the severity of each event and its relation to GH therapy. Furthermore, we evaluated adverse events developing during treatment (AEDDT), defined as new events or pre-existing events, the intensity of which increased after initiation of GH treatment, in all patients treated with GH (whether or not they had received GH prior to enrolment) that attended at least one follow-up visit. We paid especial attention to specific neoplastic events (included in the MedDRA system organ class “neoplasms benign, malignant and unspecified [incl cysts and polyps]”) and glucose metabolism disorders.

We have expressed the data as means and standard deviations (SDs) or medians and interquartile ranges (IQRs) for continuous variables, and as absolute frequencies and percentages for discrete variables. The estimated means are presented with the corresponding 95% confidence intervals (CIs). We performed the statistical analysis by means of the Statistical Analysis System software version 9.1(SAS®; SAS Institute Inc, Cary, NC, USA).

Spain participated in the GeNeSIS study with 1294 children (44% female). At the time of enrolment, 85% of the patients had never been treated with GH and 14% were already receiving it. We excluded 27 patients from the analysis because they did not receive GH therapy or due to insufficient data. Table 1 summarises the diagnoses for which GH therapy was prescribed in the 1267 patients with complete data. Most of the children had GHD (78.1%), with most cases corresponding to the classical idiopathic form of GDH (65.3% of the total treated with GH). The second most frequent category was SGA (10.9%), followed by SHOX deficiency (5.8%), of which most cases corresponded to girls with Turner syndrome.

The mean age at the time of entry in the GeNeSIS study was 9.8 years (95% CI, 9.7–10.0 years). The age interval that contained the highest proportion of patients was the 10–15 years group (45.2%), followed by the 5–10 years group (38.0%), both with similar proportions of both sexes. There was a slightly higher proportion of girls (16.5%) compared to boys (11.4%) in the group under five years, while there was a higher proportion of males (2.4%) compared to females (0.7%) in the group 15 years and older. Most patients were prepubertal (Tanner stage 1, Fig. 1) when they started treatment. At the time of the last visit, 57.6% of males and 59.3% of females were pubertal (Tanner stage ≥3).

Figure 1.

(a) Stage of pubertal development of boys during the enrolment and final visits of the GeNeSIS study in Spain. (b) Stage of pubertal development of girls during the enrolment and final visits of the GeNeSIS study in Spain. Pubertal development was assessed based on breast development (B) and genital development (G) applying the Tanner scale.

(0.15MB).

The median dose of GH at the time of entry was 0.22mg/kg/week (IQR, 0.20–0.25); after one year, it was maintained at 0.23 (IQR, 0.21–0.25) mg/kg/week, with no evident changes observed during treatment. The median initial dose was within the same range for all the diagnoses, except for Turner syndrome (median, 0.32; IQR, 0.29–0.35mg/kg/week). The median duration of followup was 2.4 years (IQR, 1.0–4.0); and followup lasted more than four years in 326 patients (25.7%). The maximum duration of followup was 10.5 years. Since 14% of patients were already being treated with GH at the time of enrolment, the median total duration of treatment exceeded 2.8 years (IQR, 1.6–4.4 years). The records indicated that 348 patients discontinued treatment. Aside from having reached their final heights or achieved a predefined height, growth velocity or bone age, the reported reasons for discontinuation were: decision by physician in 35 patients, decision by patient/parents in 32, ineffectiveness in 15, and adverse events in 2 patients.

We analysed the auxological data of 262 patients with GHD (167 male and 95 female) that had not been treated with GH prior to enrolment in the study and underwent treatment with GH for at least four years. Table 2 presents their baseline characteristics.

The GV increased from a mean of 4.3cm/year (95% CI, 4.1–4.6) at the beginning of followup to 9.0cm/year (95% CI, 8.7–9.4) by the end of the first year; reaching a value of 5.5cm/year (95% CI, 5.2–5.8) in the fourth year. The mean GV SDS (Fig. 2) increased from a value below the population mean at the beginning of followup (−1.3; 95% CI, −1.5 to −1.1) to a value above the population mean in the first year (2.9; 95% CI, 2.6–3.2), remaining high through year four (0.8; 95% CI, 0.5–1.0). The mean increase in height SDS was 0.64 (95% CI, 0.58–0.70) in the first year, and remained positive through year four (0.15; 95% CI, 0.11–0.19) (Fig. 2). The difference between the final height SDS and the target height SDS increased compared to the baseline (−1.51; 95% CI, −1.63 to −1.39) during the first year (−0.89; 95% CI, −0.99 to −0.78) and through year four (−0.21; 95% CI, −0.32 to −0.10).

Figure 2.

Growth velocity (GV) SDS and height SDS for 262 Spanish patients (167 male, 95 female) with growth hormone (GH) deficiency, who had not received prior treatment with GH, at baseline and during the four years of GH treatment. The data are expressed as means with 95% confidence intervals.

(0.08MB).

We also assessed the impact of GH treatment in the final height of 253 patients with GHD (117 male and 136 female) that were or were not being treated with GH at the time of enrolment. Table 2 shows the baseline characteristics and Table 3 the final height data for these patients. The final height SDS in these patients was close to their target height SDS, with a mean deficit of 0.09 (SD, 0.85).

The safety analysis included 1284 children (704 male, 573 female, 7 of unspecified sex) that were treated with GH during the study. The mean duration of treatment was 2.5 years (IQR, 1.0–4.1). Of the 1143 patients that had attended at least one follow-up visit, a total of 121 AEDDTs were reported in 93 children (8.1%) (Table 4); the investigators considered that 25 AEDDTs in 23 children (2.0%) were probably related to GH use. Specific neoplastic AEDDTs were reported in 7 patients and considered unrelated to treatment with GH with the exception of a case of melanocytic nevus diagnosed in a patient aged 8.4 years and treated for 0.3 years, an event that was not considered serious.

At least one SAE was reported in seven patients (0.6%) (Table 4). The investigators considered that two were at least probably related to treatment with GH: (1) one neoplastic vascular malformation in the left-eye choroid that could be fully resected in a girl aged 13.5 years with idiopathic isolated GHD that had been treated with GH for 0.9 years; and (2) an epiphysiolysis in a boy aged 14.6 years with multiple pituitary hormone deficiencies, including GHD, with a diagnosis of ectopic posterior pituitary.

Treatment was discontinued due to adverse events in two patients: (1) a boy aged 11.9 years with GHD secondary to medulloblastoma in whom a bone marrow neoplasm and obstruction of a ventriculoperitoneal shunt were suspected, and treatment discontinued after 1.6 years; both events were classified as serious and unrelated to GH treatment; and (2) male patient aged 16.5 years with idiopathic GHD in whom treatment with GH was discontinued after 3.4 years upon receiving a diagnosis of insulin resistance believed to be unrelated to treatment.