ESPE Abstracts

Background: Craniopharyngioma is a non-malignant embryonic tumour in the pituitary-hypothalamic area, associated with hypothalamic obesity. Dysfunctional parasympathetic nervous system activity has been proposed as one mechanism underlying alterations in energy metabolism. Arterial spin labelling (ASL) is a non-invasive MRI technique that quantifies brain tissue perfusion as a proxy for functional activity. Here, we measure cerebral perfusion in patients with craniopharyngioma to investigate the impact of the tumour and associated treatment on appetitive brain function, and the potential relationship with resting metabolic rate (RMR). This research forms part of a feasibility study exploring neuroimaging, eating behaviours, energy homeostasis and obesity in patients with childhood-onset craniopharyngioma.

Methods: Following an overnight fast, RMR was measured using indirect calorimetry (COSMED K4b2) and participants subsequently underwent a pseudo-Continuous ASL scan at 3T (PCASL) with multiple post-labelling delays. ASL data was processed and analysed using FSL tools in seven a priori regions of interest (ROIs). Data are reported as mean ± standard deviation. Kendall’s Tau is reported for correlations due to the small sample size of this feasibility study.

Results: Nine patients with childhood-onset craniopharyngioma (age=14.6±3.8y; BMI SDS=0.97±1.93, 3 female) and nine sex-matched historical controls (age=22.4±2.7y; BMI SDS=-0.60±0.88, 3 female) were included in the pCASL analysis. Post-operatively all patients had endocrine dysfunction; eight in thyroid axis, four with diabetes insipidus and all had adrenocorticotropic hormone and growth hormone deficiencies. Rates of perfusion were in line with the diagnostic cut-off for low-grade paediatric brain tumours (50mL/min/100g). Lower perfusion was found in the craniopharyngioma patients compared to the controls in six of the seven ROIs (P<0.01), namely hypothalamus, insula, amygdala, nucleus accumbens, putamen and ventrofrontal cortex (VFC), but not the temporal occipital fusiform cortex (TOFC) which is a visual attention region. Negative correlations between perfusion and RMR in craniopharyngioma patients were found in all ROIs (insula τ=-0.556; amygdala τ=-0.667; nucleus accumbens τ=-0.444; putamen τ=-0.500; TOFC τ=-0.667; VFC τ=-0.389), except the hypothalamus (τ=0.056).

Conclusions: Preliminary evidence suggests that brain perfusion, a proxy for neural activity at rest, is lower in craniopharyngioma patients in the appetitive brain network. Novel analyses suggest a relationship between cerebral perfusion and RMR, which may help to understand the impact of brain changes on altered energy metabolism in craniopharyngioma. The impact of these brain changes on the development of obesity should be investigated in future research.