This review presents an update on current concepts of pathogenesis, diagnostics, multidisciplinary treatment and follow-up care, with special focus on neuropsychological sequelae of childhood-onset craniopharyngioma (CP) based on most recent publications on these topics. Recent insight in molecular pathogenesis of CP opens new perspectives on targeted therapy. Further research to elucidate pathogenic mechanisms and to prevent hypothalamic involvement of CP is warranted. Surgical treatment strategies should be based on a multidisciplinary approach involving experienced teams aiming at posterior hypothalamus-sparing treatment for prevention of quality of life impairments. Centralization of CP treatment in experienced ‘centers of excellence’ is recommended. However, such centralization includes high thresholds concerning infrastructure not achievable in all health systems. Alternatives such as multicenter-based networks used for reference assessments should be considered to assure high standards of treatment quality. Irradiation is efficient in preventing further growth or recurrence in CP patients with residual tumor. Proton beam therapy – available on a wider range in the near future – will help to avoid radiooncological side effects. Novel insights into neuropsychological sequelae after CP should be the basis for the development of future therapeutic neuropsychological interventions. Due to the rareness of the disease, common international efforts in research and treatment are recommended and should lead to an international registry for childhood-onset CP, as a first step toward efficient coordination of scientific and clinical initiatives.
Svenja Boekhoff, Agnieszka Bogusz, Anthe S Sterkenburg, Maria Eveslage and Hermann L Müller
Quality of survival, prognosis and long-term outcome are often severely impaired in childhood-onset craniopharyngioma (CP) patients. Identification of risk factors for sequelae such as growth hormone (GH) deficiency is important for appropriate treatment and rehabilitation.
In a cross-sectional study, 79 CP patients recruited in HIT-Endo before 2000 were analyzed according to GH substitution: (a) CP never GH treated (noGH); (b) CP GH treated only during childhood (pedGH); (c) CP under GH, initiated at adulthood (adultGH); (d) CP under GH during childhood and continued during adulthood (contGH).
Progression-free (PFS) and overall survival (OS), height, BMI, psychosocial and neuropsychological status (EORTC QLQ-C30, MFI-20).
OS and PFS rates were similar in all subgroups. ContGH and pedGH CP presented with increases in height (P = 0.002; P = 0.0001) during long-term follow-up when compared with baseline. In all subgroups except for pedGH, increases in BMI were observed when compared with BMI at diagnosis. For emotional functionality and physical fatigue, adultGH CP showed worse (P = 0.037; P = 0.034) response (mean: 61.4%; 12.5%) when compared with pedGH CP (mean: 83.5%; 7.7%). Observed differences were not related to irradiation and hypothalamic involvement. In terms of psychosocial status, no differences were observed between subgroups.
We conclude that GH substitution was safe with regard to risk of tumor progression/relapse in CP. Growth was improved by GH, whereas the development of obesity was not influenced by GH substitution. However, early initiation of GH substitution after CP diagnosis might have beneficial effects on weight development and neuropsychological outcome.
Anika Hoffmann, Klaus Bootsveld, Ursel Gebhardt, Anna M M Daubenbüchel, Anthe S Sterkenburg and Hermann L Müller
Hypothalamic obesity in childhood craniopharyngioma (CP) patients carries a high risk for development of metabolic syndrome. In metabolic syndrome, the development of nonalcoholic fatty liver disease (NAFLD) is known. The aim of this study is to detect the risk for NAFLD in childhood-onset CP.
This cross-sectional study included liver computed tomography (CT); ultrasound analysis of abdomen; measurements of serum parameters, height, weight and body composition; and daily medication of patients with childhood-onset CP.
A total of 384 patients recruited in trials HIT Endo and KRANIOPHARYNGEOM 2000 were analyzed. Ninety-four survivors were included by fulfilling the criteria of proven hypothalamic involvement (HI), a minimum time interval of 5 years between diagnosis and study, and a minimum age of 18 years at the time of evaluation. A total of 19 patients agreed to participate. To quantify the degree of steatosis hepatis, analyses of liver density were performed once by non-contrasted CT of liver sections.
NAFLD occurs in about 50% of CP patients with HI and is associated with elevated liver enzymes and homeostasis model assessment index. BMI is not an effective predictive factor but body fat mass measured by near-infrared spectroscopy (NIRS) is. Over half of CP patients (60%) with NAFLD are treated with stimulating agents, with risk of hepatic side effects.
NAFLD is a major adverse late effect in childhood-onset CP. NIRS rather than BMI should be used to measure body composition and predict NAFLD. Stimulating agents for treatment of fatigue and daytime sleepiness in CP should be prescribed judiciously.
Anika Hoffmann, Svenja Boekhoff, Ursel Gebhardt, Anthe S Sterkenburg, Anna M M Daubenbüchel, Maria Eveslage and Hermann L Müller
Childhood craniopharyngiomas (CP) are often diagnosed after a long duration of history (DOH). Tumor size, hypothalamic involvement (HI), and obesity are associated with reduced overall survival (OS) and functional capacity (FC). The effect of DOH and specific symptoms in history on presentation at initial diagnosis and long-term prognosis are unknown.
Retrospective analysis of patients' records and prospective longitudinal follow-up.
Histories of 411 CP patients recruited in HIT Endo, KRANIOPHARYNGEOM 2000 were retrospectively evaluated for DOH, symptoms, and characteristics. The effect of specific manifestations and DOH on clinical presentation and tumor characteristics at time of initial CP diagnosis and long-term outcome were analyzed. Main outcome measures were 10-year OS and progression-free survival (PFS), FC, and BMI during longitudinal follow-up.
Median DOH was 6 months (range: 0.1–108 months) and correlated with age at diagnosis. Tumor size, HI, degree of resection, and BMI at diagnosis were not related to DOH. In multivariate analysis adjusted for age at diagnosis, only hydrocephalus was found to have a relevant influence on DOH. Visual and neurological deficits were associated with larger initial tumor size and impaired 10-year OS. Weight gain and growth failure were observed with longest DOH. PFS and FC were not related to any specific symptom. Endocrine deficits at diagnosis were associated with long DOH.
CP is frequently diagnosed after long DOH, especially in older children. However, DOH was not associated with tumor size, HI, survival, or FC. Visual and neurological deficits necessitate rapid diagnostic workup.
Hermann L Müller, Ursel Gebhardt, Carmen Teske, Andreas Faldum, Isabella Zwiener, Monika Warmuth-Metz, Torsten Pietsch, Fabian Pohl, Niels Sörensen and Gabriele Calaminus
Hypothalamic obesity has major impact on prognosis and quality of life (QoL) in childhood craniopharyngioma.
Patients and methods
For this study, 120 patients were prospectively recruited during 2001 and 2007 and evaluated after 3 years of follow-up (KRANIOPHARYNGEOM 2000). Body mass index (BMI) and QoL at diagnosis and 36 months after diagnosis were analysed based on the reference assessment of tumour localisation and post-surgical hypothalamic lesions. Treatment was analysed based on the neurosurgical strategy of 50 participating neurosurgical centres, the centre size based on the patient load.
BMI SDS at diagnosis was similar in patients with or without hypothalamic involvement. Surgical lesions of anterior and posterior hypothalamic areas were associated with higher increase in BMI SDS during 36 months post-diagnosis compared with patients without or only anterior lesion (+1.8 BMISD, P=0.033, +2.1 BMISD; P=0.011), negative impact on QoL in patients with posterior hypothalamic lesions. Surgical strategies varied among the 50 neurosurgical centres (three large-sized, 24 middle-sized and 23 small-sized centres). Patients treated in small-sized centres presented with a higher rate of hypothalamic involvement compared with those treated in the middle- and large-sized centres. Treatment in large-sized centres was less radical, and the rates of complete resection and hypothalamic surgical lesions were lower in large-sized centres than those of the middle- and small-sized centres. However, a multivariable analysis showed that pre-operative hypothalamic involvement was the only independent risk factor for severe obesity (P=0.002).
Radical neurosurgical strategies leading to posterior hypothalamic lesions are not recommended due to the potential to exacerbate hypothalamic obesity and impaired QoL. Treatment should be confined to experienced multidisciplinary teams.