ESPE2021 ePoster Category 1 Fat, Metabolism and Obesity B (10 abstracts)
1NIHR Bristol Biomedical Research Centre Nutrition Theme, University of Bristol, Bristol, United Kingdom; 2Department of Paediatric Endocrinology & Diabetes, Bristol Royal Hospital for Children, University Hospitals Bristol NHS Foundation Trust, Bristol, United Kingdom; 3Bristol Medical School, University of Bristol, Bristol, United Kingdom
Background: Craniopharyngioma is a rare, suprasellar tumour, which, together with associated surgery or radiotherapy, results in damage to the hypothalamus and severe obesity in approximately 50% of cases. The multi-factorial mechanisms underlying the development of obesity in craniopharyngioma are not well understood. Hypothalamic damage in craniopharyngioma has been associated with dysfunctional parasympathetic nervous system activity leading to altered glucose metabolism and hyperinsulinemia (van Iersel et al., Endocrine Reviews, 2019). Furthermore, additional evidence in lean and obese volunteers suggests a link between insulin resistance, neural responses to food and obesity (Cheke et al., Neuropsychologia, 2017). We therefore hypothesised that insulin resistance in craniopharyngioma may alter neural response to food cues and thereby contribute to obesity.
Method: A feasibility study of 11 young craniopharyngioma patients (median [range] age = 14 [10-23]y; BMISDS = 1.3 [-1.5-3.9]; 5F/6M) with full clinical characterisation was conducted, including measurement of body composition using bioelectrical impedance and an oral glucose tolerance test (n = 10); during which preliminary functional neuroimaging data on the neural response to food cues, in a priori regions of interest, during fasting and post-glucose ingestion were collected (n = 9). HOMA-IR was calculated to assess insulin resistance. As this feasibility study was not powered for null hypothesis significance testing, we focussed on measure of effect size (Kendalls Tau correlation coefficient).
Results: Normal glucose tolerance (based on WHO criteria) was found (median fasting glucose = 4.45 mmol/l) in all but one patient, who had Type 2 diabetes (removed from HOMA-IR correlational analyses below as an outlier). Four patients were classed as insulin resistant, with a HOMA-IR of > 2.5 (median HOMA-IR = 3.11). No evidence for a correlation was found between HOMA-IR and body fat % (τ = -.25), fat mass (kg) (τ = -.22,) or BMISDS (τ = .00). Preliminary evidence was found for a positive correlation between HOMA-IR and (i) fasting response to food cues in the insula (τ = .43) and the putamen (τ = .52), and (ii) post-glucose response to food cues in the amygdala (τ = .62) and nucleus accumbens (τ = .43). By contrast, BMISDS was negatively correlated with the neural response in the hypothalamus during fasting (τ = .56) and post-glucose (τ = -.56).
Conclusion: These preliminary findings in this small group of craniopharyngioma patients suggest that insulin resistance may influence the neural response to food cues before and after glucose consumption in reward regions of the brain, different to those regions influenced by body mass.
Trial registered with ISRCTN (isrctn.com/ISRCTN86005167).