Septo-optic dysplasia/optic nerve hypoplasia: data from the National Cooperative Growth Study (NCGS)

We analyzed data from 65 children with septo-optic dysplasia (SOD) referred for evaluation and followed in the National Cooperative Growth Study (NCGS) Substudy 8 and from 758 children treated with growth hormone (GH) and followed in the NCGS core study. Compared to other children referred for evaluation of short stature, children with SOD were younger (mean age 3.7 +/- 3.6 vs 8.6 +/- 4.9 years), had less severe short stature (mean +/- SD height SDS -1.80 +/- 1.64 vs -2.17 +/- 0.95), and were more likely to be female (46% F vs 31% M). Children with SOD who received GH were older and shorter than those referred and untreated, but the gender distribution was similar. Other pituitary hormone deficits were reported in untreated patients, including thyroid hormone deficiencies (8%) and adrenocorticotropic hormone (ACTH) deficiency (3%), as compared to 27% and 24%, respectively, in GH-treated children. Data on adult height were available for 71 patients, who showed an average gain in height SDS of 1.17 +/- 1.49. GH therapy was well tolerated in children with SOD.     

Intracranial hypertension in pediatric patients treated with recombinant human growth hormone: data from 25 years of the Genentech National Cooperative Growth Study

Intracranial hypertension (IH) is a rare condition in children. However, a relationship between recombinant human growth hormone (rhGH) therapy and IH has been well documented. Risk factors were assessed for 70 rhGH-naive patients enrolled in the National Cooperative Growth Study with reports of IH after treatment initiation. Patients with severe growth hormone deficiency, Turner syndrome, chronic renal insufficiency (CRI), and obesity (particularly in the CRI group) were at highest risk of developing IH during the first year of therapy, suggesting initiation of careful early monitoring. In some patients, factors such as corticosteroid use or other chromosomal abnormalities appear to confer a delayed risk of IH, and these patients should be monitored long-term for signs and symptoms of IH.     

Growth hormone safety update from the National Cooperative Growth Study

We reviewed adverse event (AE) data in the National Cooperative Growth Study from start-up (1985) until January 1, 1999. Enrollment was 33,161. A total of 2,632 AE reports were received; 863 were serious events, with 156 deaths. The most common cause of death was recurrence of intracranial neoplasm. There were 20 reports of leukemia, and the standard morbidity ratio (SMR) was 0.73 (95% CI: 0.20-1.86) for the four cases without risk factors. There were 35 reports of extracranial nonleukemic malignancy, and the SMR was 0.44 (95% CI: 0.24-0.74) for the 14 cases without risk factors. The recurrence rate for all brain tumors present at baseline was 7.6%, and for craniopharyngiomas, 6.4%. There were 49 reports of intracranial hypertension (20 patients had papilledema), 68 reports of diabetes/hyperglycemia, 45 of slipped capital femoral epiphysis, 136 of scoliosis, and five of pancreatitis. There was no evidence of increased incidence of leukemia or extracranial nonleukemic malignancies among patients without prior risk factors. Intracranial hypertension does not necessarily occur early in growth hormone therapy. Other findings were consistent with past observations.     

Use of magnetic resonance imaging in short stature: data from National Cooperative Growth Study (NCGS) Substudy 8

The primary use of magnetic resonance imaging (MRI) in the evaluation of children with short stature (SS) is to discover lesions in the central nervous system (CNS), particularly tumors that may require intervention. MRI has a secondary role in identifying structural abnormalities responsible for growth hormone deficiency (GHD). We examined data from the National Cooperative Growth Study (NCGS) Substudy 8 to determine how American physicians are using MRI in evaluating children with SS. Of the 21,738 short children enrolled in NCGS, 5% underwent MRI during their follow-up. Children who had GH stimulation testing were more likely to have had an MRI than those in whom no GH stimulation test was performed (19% vs 2%, p <0.0001). Moreover, children diagnosed with severe GHD (maximum GH <5 ng/ml) were more likely to have an abnormal finding on MRI. Of these patients, 27% demonstrated an abnormality as compared to 12% and 12.5% in patients with partial GHD and normal GH stimulation test results (>10 ng/ml), respectively. Abnormalities unrelated to the hypothalamus or pituitary represented 30% of these findings, while disorders in pituitary anatomy, including pituitary hypoplasia, pituitary stalk interruption, and ectopic posterior pituitary, represented an additional 30% of abnormal MRI examinations. CNS tumors comprised 23% of abnormal findings in these patients. We conclude that MRI provides significant value in the evaluation of children with SS, by identifying CNS tumors associated with growth failure as well as anatomical abnormalities of the pituitary. These findings are useful in confirming the diagnosis of GHD in children and identifying potential candidates for continued GH replacement in adulthood.     

Growth hormone responsiveness: peak stimulated growth hormone levels and other variables in idiopathic short stature (ISS): data from the National Cooperative Growth Study

HYPOTHESIS: In children with idiopathic short stature (ISS), growth hormone (GH) response to a provocative test will be inversely related to the first year response to hGH and be a variable accounting for a degree of responsiveness. BACKGROUND: Because high levels of GH are a characteristic of GH insensitivity, such as in Laron syndrome, it is possible that a high stimulated GH is associated with a lower first year height velocity among children diagnosed as having ISS. METHODS: We examined the relationship between the peak stimulated GH levels in 3 ISS groups; GH >10 -<25, 25-40, and >40 ng/mL and the first year growth response to rhGH therapy. We also looked at 8 other predictor variables (age, sex, height SDS, height age, body mass index (BMI), bone age, dose, and SDS deficit from target parental height. Multiple regression analysis with the first year height as the dependent variable and peak stimulated GH was the primary endpoint. The predictive value of adding each of the other variables was then assessed. RESULTS: Mean change in height velocity was similar among the three groups, with a maximum difference among the groups of 0.6 cm/yr. There was a small but statistically significant correlation (r=-0.12) between the stimulated GH and first year height velocity. CONCLUSIONS: The small correlation between first year growth response and peak GH is not clinically relevant in defining GH resistance. No cut off level by peak GH could be determined to enhance the usefulness of this measure to predict response. Baseline age was the only clinically significant predictor, R-squared, 6.4%. All other variables contributed less than an additional 2% to the R-squared.     

Effect of long-term recombinant growth hormone therapy in children--the National Cooperative Growth Study, USA, 1985-1994

Over a 9-year period (1985-1994) approximately 20,000 children received recombinant human growth hormone (rhGH) while enrolled in the National Cooperative Growth Study (NCGS), an observational, longitudinal study designed to monitor the long term efficacy and safety of rhGH administered to children in North America. Forty-four percent of the patients had idiopathic growth hormone deficiency (IGHD), 13.8% organic GHD (OGHD), 25% idiopathic short stature (ISS), 9.9% Turner's syndrome (TS), and 7.3% miscellaneous disorders. Eighty-five percent of the patients enrolled were Caucasian, and approximately two-thirds of the non-Turner patients were male. For the subset of patients treated for at least 4 years and who were prepubertal throughout this period (IGHD N=308, OGHD N=93, ISS N=169, TS N=82), mean growth rates increased in all patient categories and remained at or above pretreatment growth rates through 4 consecutive years of therapy with rhGH. Growth rates during administration of rhGH were greater in children in whom the pretreatment maximum stimulated GH concentration was < or =3 microg/l. Patients treated with 6 or 7 doses of rhGH each week grew more rapidly than did those receiving thrice weekly dosages, although the ratios of the increment in bone age to the increment in height age after two years of therapy were similar in the two treatment regimens. For patients treated with rhGH for 7 consecutive years, the mean height standard deviation scores increased by 2.5 in IGHD (N=169), 2.0 in OGHD (N=50), 1.9 in ISS (N=69), and 1.3 in TS (N=19), but remained below target heights in all categories. It is concluded that administration of rhGH increases growth rates in patients with IGHD, OGHD, ISS, and TS, and that this stimulatory effect can persist for at least 4 years.     

Growth impairment and growth hormone therapy in children treated for malignant brain tumours

Eighty-two children with malignant brain tumours were treated according to the “8 in 1” chemotherapy protocol in Finland during 1986 to 1993. Thirty-seven with brain tumours not involving the hypothalamic-pituitary region are still alive and tumour-free. The growth and response to growth hormone (GH) therapy in these children was analysed. Children who received craniospinal irradiation had the most severe loss of height SDS, being −1.07 within 3 years of the diagnosis. Even children with no irradiation to the hypothalamic-pituitary axis had a mean change in height SDS of −0.5 after 3 years. Fifteen of 23 children who received craniospinal irradiation and two out of eight children who received cranial irradiation have received GH therapy. A catch-up growth response to the daily GH therapy with the mean dose of 0.7 IU/kg per week was complete in 3 years (+1.87 SDS), irrespective of craniospinal irradiation, in children who were treated at prepubertal age but was seen in none of the children who had reached pubertal age. Conclusion Growth impairment and GH deficiency are common in children treated for malignant brain tumours. The response to GH therapy is good in prepubertal children in terms of increased growth velocity, although the final height is not yet known. Received: 10 September 1996 and in revised form: 28 January 1997 / Accepted: 11 February 1997     

A longitudinal study on growth and spontaneous growth hormone (GH) secretion in children with irradiated brain tumors

Longitudinal growth was studied in 27 children after radiotherapy for a brain tumor. Growth deviation (greater than or equal to 1 SD) was found in 56% of the children after 2 years and was most profound in prepubertal children aged between 3 and 8 years at the time of irradiation. In this group growth velocity was markedly reduced and no catch up was seen. In all children studied growth hormone (GH) secretion, measured as the spontaneous secretion over 24 hours, was found to be severely disturbed. Our conclusion is that all children with a growth deviation greater than or equal to 1 SD after radiotherapy (greater than or equal to 40 Gy) to the hypothalamo-hypophyseal region should be considered GH deficient. In such children GH treatment can be initiated without further testing.     

Auxologic and biochemical characterization of the three phases of growth failure in pediatric patients with brain tumors

Pediatric patients with brain tumors can loose 1 SD of height prior to beginning growth hormone (GH) therapy. The objectives of this study were to characterize the early growth failure, identify contributing factors and propose interventions. Five children were followed quarterly for 2 years to monitor auxological parameters, nutritional indices, and endocrine measuremnts. GH stimulation tests were done every 6 months to determine the timing of the onset of GH deficiency. The nadir for height velocity (HV) occurred 6 months after diagnosis. Poor gains in height correlated with decreased calorie count (p <0.001), poor weight gain (p <0.001), decreased BMI (p <0.001) and lowered leptin levels (p <0.001). All patients were able to secrete GH normally during this nadir of growth. Children treated for brain tumors demonstrate an early triphasic pattern of growth. Growth failure due to cachexia occurs first, then a second transient phase of normal growth is observed followed by a third phase of growth failure due to GH deficiency. Phase 1 is characterized by decreased HV, BMI, leptin levels and calorie counts. With recognition of this profile, the early growth failure might be preventable with aggressive nutritional rehabilitation.     

Recurrence of brain tumours in patients treated with growth hormone: analysis of KIGS (Pfizer International Growth Database)

Background: Growth hormone (GH) has been used successfully in the treatment of short stature secondary to GH deficiency in survivors of childhood brain tumours. There has been concern that GH might increase the risk of recurrence. Aim: To analyse KIGS (Pfizer International Growth Database) with respect to tumour recurrence in patients with brain tumours. Methods: Data for tumour recurrence were analysed retrospectively in 1038 patients with craniopharyngiomas, 655 with medulloblastomas, 113 with ependymomas, 297 with germinomas, and 400 with astrocytomas or gliomas. All patients had received recombinant human GH (Genotropin®, Pfizer Inc.). Results: Recurrence-free survival rates were 63% at a follow-up of 10.3 y in craniopharyngioma, 69% in 9.1 y in the glial tumours, 71% in 7.4 y in germinomas, 92% in 4.6 y in medulloblastomas and 89% in 2.5 y in ependymomas. Dose of GH and treatment modalities did not differ significantly between patients with and without recurrence.
Conclusion: Tumour recurrence rates in surviving patients with brain tumours receiving GH treatment do not appear to be increased compared with published reports. However, longer follow-up regarding recurrences and secondary neoplasms remains essential.     

Risk of venous thromboembolism in pediatric patients with brain tumors

BACKGROUND: Venous thromboembolism (VTE) is a common event in adults with malignant brain tumors approaching 24% throughout the course of the disease. The high morbidity and mortality of this complication yielded several protocols for prevention of the disease in adults undergoing neurosurgery for brain tumors and possible primary prevention afterwards. We investigated the incidence and complications of VTE in pediatric neuro-oncology patients. PROCEDURE: We analyzed, retrospectively, the files of all consecutive patients under the age of 18 years who were hospitalized for the treatment of brain tumors between the years 1990 and 2003 in two leading, closely related, Israeli neuro-oncology centers. RESULTS: A total of 462 children were analyzed. Three hundred eighty-four patients underwent surgery and 78 were treated medically. Only three (0.64%) of the patients developed clinical episodes of VTE that were treated conservatively. Two of these patients developed intracranial bleeding while on secondary prevention for the disease. CONCLUSIONS: Although this study has considerable limitations in terms of retrospective design, heterogeneous group of patients and diagnoses, the changing awareness for thrombosis over the last 14 years and the inclusion of symptomatic VTE events only, our surprising data suggest that, as opposed to adults, the risk of clinically significant VTE in children with brain tumors may be exceedingly low. These findings set the stage for future forthcoming evaluations in view of the prospective studies that were done in adults and the possible significant implications for the prevention and possible etiologies of the disease.     

Pathologically proven cavernous angiomas of the brain following radiation therapy for pediatric brain tumors

Lesions consistent with cavernous angiomas (CAs) of the brain are sometimes seen on MRI scans of the brains of patients who received radiation therapy for brain tumors as children. The lesions appear years later within brain tissue that was included in radiation fields. It is unclear whether these MRI-detected lesions are true CAs or a pathological variant. This study reports the clinical, radiographical, and pathological findings in 3 cases of radiation-induced CAs of the brain. From 1995 to 1997, 3 patients previously treated with radiation therapy (45-55 Gy) for pediatric brain tumors (medulloblastoma, ependymoma, and a presumed midbrain astrocytoma) underwent resections of symptomatic and enlarging lesions that were consistent with a CA of the brain. All of the lesions occurred within fields of prior irradiation. None of the patients had received chemotherapy as part of their cancer treatment. CA-presenting symptoms included seizures, cranial nerve deficits, and headaches. The lesions appeared 7-19 years after radiation therapy and slowly enlarged on subsequent imaging studies. MRI scans of the lesions revealed characteristics typical of CA. The lesions became symptomatic 1-5 years after they were initially noted. Surgical resection was performed 1-2 years after symptoms began. The age at resection ranged from 15 to 23 years (10-21 years after radiation therapy). Pathological analysis of the three lesions showed typical CA characteristics. Some CAs may be caused by radiation therapy for pediatric brain tumors. They are radiologically and pathologically similar to sporadically occurring CAs of the brain and may enlarge over time and become symptomatic. CAs can be safely resected using standard microsurgical techniques.     

Intracranial endodermal sinus tumors associated with growth hormone replacement therapy in a girl

The primary intracranial endodermal sinus tumor (EST) is regarded as a rare histological subtype that is often associated with components of other germ cell tumors, and there are no reports on the onset of intracranial ESTs after growth hormone (GH) replacement therapy. The authors report an extremely rare case of pure primary EST associated with GH replacement therapy. A 15-year-old girl with GH deficiency experienced headache, nausea, and vomiting after GH replacement therapy for a 17-month period. Magnetic resonance imaging showed 2 tumor masses located in the pineal region and frontal horn of the right lateral ventricle, respectively. Before surgery, the authors administered 1 cycle of neoadjuvant chemotherapy, which shrank the tumor and facilitated surgical intervention. The larger mass located in the pineal region was removed via a right occipital transtentorial approach, and postoperative histopathological analysis revealed a pure EST. While there is a clear association between the initiation of GH replacement therapy and the development of the EST in this case, the causal effect cannot be specified. Nevertheless, this case demonstrates that GH replacement therapy must be used cautiously.     

New technologies in radiation therapy for pediatric brain tumors: the rationale for proton radiation therapy

BACKGROUND: Pediatric brain tumors are frequently treated with radiation therapy and often cured. The long-term side effects of treatment with high-energy X-rays (photons) can be substantial. Proton radiation therapy may limit these late effects. PROCEDURE: The physical difference between photon and proton irradiation is compared. The clinical benefits of the superior physical properties of proton beam radiation therapy are explained for children with brain tumors. RESULTS: At biologically equivalent doses, proton radiation therapy offers tumor control similar to photon radiation therapy. The superior physical properties of proton beams make this mode of radiation therapy less likely to cause late effects. CONCLUSIONS: For many children with brain tumors, proton beam radiation therapy may limit the late effects of radiation therapy and therefore offer an advantage over techniques using photons.     

Short- and long-term complications of radiation therapy for pediatric brain tumors.

Brain tumors are the second most common malignancy of childhood after acute lymphocytic leukemia. Improvements in therapy have led to increased survival. It is estimated that by the year 2000 there will be approximately 200,000 survivors of childhood cancer. A proportion of these will have survived a central nervous system malignancy. As more children survive, complications of treatment are increasingly recognized. This paper reviews the spectrum of radiation-induced complications, both short- and long-term. Their frequency and characteristics will be reviewed as well as suggestions made to decrease their incidence.