Antibodies Against Hypothalamus and Pituitary Gland in Childhood-Onset Brain Tumors and Pituitary Dysfunction.

Purpose: To detect the presence of antipituitary (APA) and antihypothalamus antibodies (AHA) in subjects treated for brain cancers, and to evaluate their potential association with pituitary dysfunction. Methods: We evaluated 63 patients with craniopharyngioma, glioma, and germinoma treated with surgery and/or radiotherapy and/or chemotherapy at a median age of 13 years. Forty-one had multiple pituitary hormone deficiencies (MPHD), six had a single pituitary defect. GH was the most common defect (65.1%), followed by AVP (61.9%), TSH (57.1%), ACTH (49.2%), and gonadotropin (38.1%). APA and AHA were evaluated by simple indirect immunofluorescence method indirect immunofluorescence in patients and in 50 healthy controls. Results: Circulating APA and/or AHA were found in 31 subjects (49.2%) and in none of the healthy controls. In particular, 25 subjects out of 31 were APA (80.6%), 26 were AHA (83.90%), and 20 were both APA and AHA (64.5%). Nine patients APA and/or AHA have craniopharyngioma (29%), seven (22.6%) have glioma, and 15 (48.4%) have germinoma. Patients with craniopharyngioma were positive for at least one antibody in 39.1% compared to 33.3% of patients with glioma and to 78.9% of those with germinoma with an analogous distribution for APA and AHA between the three tumors. The presence of APA or AHA and of both APA and AHA was significantly increased in patients with germinoma. The presence of APA (P = 0.001) and their titers (P = 0.001) was significantly associated with the type of tumor in the following order: germinomas, craniopharyngiomas, and gliomas; an analogous distribution was observed for the presence of AHA (P = 0.002) and their titers (P = 0.012). In addition, we found a significant association between radiotherapy and APA (P = 0.03). Conclusions: Brain tumors especially germinoma are associated with the development of hypothalamic-pituitary antibodies and pituitary defects. The correct interpretation of APA/AHA antibodies is essential to avoid a misdiagnosis of an autoimmune infundibulo-neurohypophysitis or pituitary hypophysitis in patients with germinoma.

Central diabetes insipidus in children and young adults: etiological diagnosis and long-term outcome of idiopathic cases.

CONTEXT: Central diabetes insipidus (CDI) is considered idiopathic in 20% to 50% of affected subjects. OBJECTIVE: The purpose of this study was to determine whether a systematic diagnostic workup could achieve better etiologic diagnosis in children and adolescents presenting with polyuria and polydipsia. DESIGN AND SETTING: This is a prospective study conducted at a tertiary referral center. Patients underwent clinical and endocrine evaluations every 6 months and neuroimaging every 6 months for 2 years and yearly for 3 years. Endocrine function and neuroimaging were also reassessed after adult height achievement. PARTICIPANTS: A total of 85 consecutive patients with CDI were enrolled at a median age of 7.5 years; those with idiopathic CDI were stratified based on pituitary stalk thickness. MAIN OUTCOME MEASURES: To establish the etiology of CDI, we determined the time lag between its onset and the specific diagnosis, the long-term impact on pituitary function, and the overall long-term outcomes. RESULTS: Of the subjects, 24 (28.2%) received an etiologic diagnosis at presentation and 11 (13%) within 2.5 years (n = 7 germinomas and n = 4 Langerhans cell histiocytosis), 7 (8.2%) were lost to follow-up, and 43 (50.6%) were considered to have idiopathic disease and were followed until the median age of 17.3 years. Neuroimaging identified 40 of 43 patients with self-limited inflammatory/autoimmune pituitary stalk thickness within the first 6 months, the severity of which was significantly correlated to pituitary dysfunction. The probability of >10-year-survival without an anterior pituitary defect was related to the severity of pituitary stalk thickness, and 53% showed permanent anterior pituitary defects. Three patients developed Langerhans cell histiocytosis and 1 developed Hodgkin lymphoma after a median of 9 and 13 years, respectively. CONCLUSIONS: A diagnostic etiology was achieved in 96% of patients with CDI. Risk stratification based on the degree of pituitary stalk thickness is of prognostic value for long-term outcomes including permanent pituitary dysfunction. New guidance is provided for the management of these patients.

The accuracy of the glucagon test compared to the insulin tolerance test in the diagnosis of adrenal insufficiency in young children with growth hormone deficiency.

CONTEXT: The accuracy of the glucagon test in the diagnosis of central adrenal insufficiency in young children has not yet been definitively established. OBJECTIVE: The aim of this study was to investigate the diagnostic accuracy of the glucagon test as an alternative to the insulin tolerance test (ITT) in children with GH deficiency under 6 yr of age. DESIGN AND SETTING: This was a prospective study conducted in two Pediatric Endocrinology Centers. PATIENTS AND METHODS: Forty-eight children (median age, 4.2 yr) with GH deficiency confirmed by a peak GH to ITT and arginine less than 10 microg/liter were enrolled: 24 with normal hypothalamic-pituitary anatomy, seven with isolated anterior pituitary hypoplasia, and 17 with structural hypothalamic-pituitary abnormalities at magnetic resonance imaging. Twelve subjects had central adrenal insufficiency defined by a peak cortisol response of less than 20 microg/dl to ITT. All children underwent a glucagon stimulation test with blood sampling for cortisol and glucose (time 0 to 180 min) after the im administration of 30 microg/kg of glucagon. RESULTS: The mean peak cortisol after glucagon was not significantly different from that obtained after ITT in the whole cohort (25.9 vs. 26.0 microg/dl; P = 0.908), and it was significantly reduced in patients with structural hypothalamic-pituitary abnormalities (P < 0.001). Receiver operating characteristic curve analysis showed that the best diagnostic accuracy was obtained with a peak cortisol cutoff to glucagon of 14.6 microg/dl (sensitivity, 66.67%; specificity, 100%; area under the curve = 0.91; 95% confidence interval, 0.82-0.99). Using this cutoff, 91.67% of the patients were correctly classified. CONCLUSIONS: This study shows that glucagon is an accurate and safe diagnostic test for adrenal function in young children who are at risk for adrenal insufficiency.

The glucagon test in the diagnosis of growth hormone deficiency in children with short stature younger than 6 years.

CONTEXT: Few studies have addressed the diagnostic role of the glucagon test in children with suspected GH deficiency (GHD). OBJECTIVE: The objective of the study was to investigate the diagnostic value of the glucagon test as an alternative test to insulin tolerance test (ITT) and arginine in GHD children younger than 6 yr. DESIGN AND SETTING: This study was conducted in two pediatric endocrinology centers. PATIENTS AND METHODS: Forty-eight children (median age 4.2 yr, median height -3.0 sd score) with GHD confirmed by a peak GH to ITT and arginine less than 10 microg/liter (median 4.7 and 3.4 microg/liter, respectively) underwent a glucagon stimulation test. Magnetic resonance imaging showed normal hypothalamic-pituitary anatomy in 24 children, isolated anterior pituitary hypoplasia in seven, and structural hypothalamic-pituitary abnormalities in 17. RESULTS: Median GH peak response to glucagon (13.5 microg/liter) was significantly higher than that observed after ITT and arginine (P < 0.0001). GH peak after glucagon was less than 10 microg/liter in 20 subjects (group 1) and greater than 10 microg/liter in 28 subjects (group 2) without significant clinical or biochemical differences between the two groups. Median GH peak after glucagon was similar between patients with multiple pituitary hormone deficiency and those with isolated GHD and between subjects with and without structural hypothalamic-pituitary abnormalities. The magnitude of the GH peak after glucagon was negatively correlated to age at diagnosis (rho = -0.636, P < 0.0001). CONCLUSIONS: This study shows that glucagon has an effective GH-releasing activity and can be used to evaluate somatotroph function in young children with short stature. Normative data for this test in young children need to be established before its use in clinical practice.

Reassessment of the growth hormone status in young adults with childhood-onset growth hormone deficiency: reappraisal of insulin tolerance testing.

CONTEXT: The 2007 Consensus Statement suggested a peak GH cutoff to insulin tolerance test (ITT) of less than 6 microg/liter in the diagnosis of permanent GH deficiency (GHD) in young adults with childhood-onset GHD (COGHD), although further validation was recommended. OBJECTIVE: The aim of the study was to evaluate the accuracy of ITT, mean 12-h spontaneous nocturnal GH (SNGH), and IGF-I in the definition of permanent GHD. DESIGN AND SETTING: The study was conducted in two Pediatric Endocrinology Centers. PATIENTS AND METHODS: ITT, 12-h SNGH, and IGF-I were evaluated as single or combined tests in 79 subjects with COGHD (median age, 18.0 yr). The cohort consisted of 48 subjects with isolated GHD or one additional pituitary defect and normal MRI or anterior pituitary hypoplasia (group LLGHD, low likelihood GHD), and 31 subjects with structural hypothalamic-pituitary abnormalities or multiple pituitary hormone deficiencies (group HLGHD, high likelihood GHD). RESULTS: Receiver operating characteristic analysis showed the best diagnostic accuracy for peak GH cutoffs to ITT of 5.62 microg/liter or less [sensitivity, 77.4%; specificity, 93.8%; area under the curve (AUC) = 0.92], mean 12-h SNGH of 1.20 microg/liter or less (sensitivity, 90.3%; specificity, 89.6%; AUC = 0.93), and IGF-I of -2.83 sd score or less (sensitivity, 80.7%; specificity, 95.7%; AUC = 0.93). Seven patients in group HLGHD showed a peak GH to ITT above 5.62 microg/liter, but a median IGF-I that was significantly lower than that of group LLGHD (-3.30 vs. -0.73 sd score; P = 0.0001). Peak GH to ITT of 3.6 microg/liter or less and arginine of 3.1 microg/liter or less at childhood diagnosis can predict a future permanent GHD condition. CONCLUSIONS: The adopted peak GH to ITT below 5.62 microg/liter is an accurate diagnostic cutoff point for HLGHD in young adults with COGHD. In addition, IGF-I is a reliable marker providing information about the severity of GHD. Careful follow-up is required for subjects with discordant ITT and IGF-I results.

Developmental abnormalities of the posterior pituitary gland.

While the molecular mechanisms of anterior pituitary development are now better understood than in the past, both in animals and in humans, little is known about the mechanisms regulating posterior pituitary development. The posterior pituitary gland is formed by the evagination of neural tissue from the floor of the third ventricle. It consists of the distal axons of the hypothalamic magnocellular neurones that shape the neurohypophysis. After its downward migration, it is encapsulated together with the ascending ectodermal cells of Rathke's pouch which form the anterior pituitary. By the end of the first trimester, this development is completed and vasopressin and oxytocin can be detected in neurohypophyseal tissue. Abnormal posterior pituitary migration such as the ectopic posterior pituitary lobe appearing at the level of median eminence or along the pituitary stalk have been reported in idiopathic GH deficiency or in subjects with HESX1, LHX4 and SOX3 gene mutations. Another intriguing feature of abnormal posterior pituitary development involves genetic forms of posterior pituitary neurodegeneration that have been reported in autosomal-dominant central diabetes insipidus and Wolfram disease. Defining the phenotype of the posterior pituitary gland can have significant clinical implications for management and counseling, as well as providing considerable insight into normal and abnormal mechanisms of posterior pituitary development in humans.