Food Preferences and Craniopharyngiomas

Hypothalamic obesity is a rare but severe condition characterized by rapid and often refractory weight gain following hypothalamic lesions. Among suprasellar tumors, craniopharyngiomas are the most frequent, and up to 40% of patients present with obesity at diagnosis. Over the course of treatment, including surgical management, the prevalence of obesity may rise to 60%. The most pronounced weight gain typically occurs within the first year after surgery. This weight increase is multifactorial, involving hypothalamic damage to nuclei regulating appetite and satiety, neuroendocrine dysfunction, reduced basal metabolic rate, decreased physical activity, and autonomic imbalance leading to hyperinsulinism. Additional contributors include neurocognitive and sleep disturbances, visual impairment, and medication side effects.

Beyond the quantitative increase in food intake, food palatability is a key determinant of dietary choices and caloric intake. Highly palatable foods, generally rich in fats and sugars, stimulate brain reward circuits, reinforcing their consumption and thereby promoting excess caloric intake and weight gain. While such mechanisms have been described in common obesity, food preferences have never been systematically investigated in patients with hypothalamic lesions, particularly those with craniopharyngiomas. Animal studies suggest that the lateral hypothalamus plays a critical role in responses to addictive substances as well as to fat and sugar consumption, raising the hypothesis that hypothalamic tumors may alter food preferences in humans and contribute to hypothalamic obesity.

This study, entitled PREFAMHYPO, is a two-phase, non-interventional observational clinical investigation designed to explore the role of food preferences in hypothalamic obesity.

Study Objectives and Design

The study has two main phases:

The objective of Phase 1 is the cultural adaptation and psychometric validation of the French version of the Food Preference Questionnaire (FPQ). The questionnaire, which assesses preferences across major food categories, has been translated and back-translated to ensure accuracy. This phase will confirm its reliability and validity in a French adult population, using measures such as internal consistency (Cronbach's alpha), test-retest reproducibility, and exploratory factor analysis. A subgroup of 50 participants will complete the FPQ twice, two weeks apart, to assess reproducibility.

Phase 2 consists of a cross-sectional evaluation of food preferences using the validated FPQ in three groups :

• Adults surgically treated for craniopharyngiomas
• Adults surgically treated for pituitary macroadenomas with at least two anterior pituitary hormonal deficits matched to craniopharyngioma patients on age, sex, and body mass index (BMI).
• Control participants without hypothalamic-pituitary disease matched to craniopharyngioma patients on age, sex, and body mass index (BMI).

The primary objective of this second phase is to compare food preferences across groups, with a specific focus on preferences of high-fat and high-sugar foods. Secondary objectives include evaluation of fat preferences, evaluation of eating behavior via the validated French version of DEBQ, evaluation of physical activity via the validated version of IPAQ, anthropometrics, body composition, and standard hormonal/metabolic panels.

Study Procedures

For phase 1, after receiving detailed information and providing documented non-opposition, participants will complete a set of standardized, self-administered questionnaires at inclusion and only 50 participants will complete the questionnaires again after 14 days.

For phase 2, after receiving detailed information and providing documented non-opposition, participants will complete a set of standardized, self-administered questionnaires at inclusion during a routine consultation or hospitalization, with no subsequent follow-up.

For patients, relevant anthropometric, clinical, and biological data will be retrieved from medical records, including BMI, waist circumference, hormone levels, metabolic parameters, treatments, comorbidities, and surgical details.

Data Management and Quality Assurance

Data will be collected by investigators at participating AP-HP sites and entered into a secure, password-protected REDCap system or encrypted Excel database. Patient anonymity will be preserved through pseudonymization, using site codes, inclusion numbers, and initials. The link between codes and identities will be kept locally under investigator control.

Source documents include medical records, consultation reports, surgical notes, laboratory results, imaging reports, and the completed questionnaires. All gathered data relevant to this study will be retained for 15 years, in accordance with French regulations. Only authorized study personnel or monitors will have access, under conditions ensuring confidentiality and compliance with Good Clinical Practice (GCP).

Data quality will be ensured through predefined validation rules, consistency checks, and central monitoring by the sponsor (AP-HP). Statistical analyses will be performed on pseudonymized datasets.

Sample Size and Statistical Analysis

The sample size for Phase 1 is based on Cattell's method, requiring at least three participants per item. With 62 FPQ items, 186 participants are needed.

For Phase 2, the rarity of craniopharyngiomas necessitates an exploratory design with 40 cases, 40 macroadenoma patients, and 80 controls. This sample is sufficient to provide descriptive and comparative analyses, generate hypotheses, and inform future larger-scale studies.

Statistical methods include

Phase 1 : Exploratory factor analysis (principal component analysis with Varimax rotation), Cronbach's alpha for internal consistency, intraclass correlation coefficients for reproducibility, and validity testing against external constructs (e.g., psychological distress measured by Kessler-10).

Phase 2 : Descriptive statistics, ANOVA or Kruskal-Wallis tests for group comparisons, post hoc multiple comparison tests, correlation analyses (Pearson or Spearman), and multivariable linear regression models to identify independent factors associated with food preferences. Missing data will be handled using appropriate statistical techniques, including imputation when necessary.

Duration

The total study duration is estimated at 36 months, including 24 months of participant recruitment and 12 months for data analysis and reporting.

Inclusion of participants for Phase 1 will be done within three months and for phase 2 within 21 months. Each participant's involvement is limited to one visit except for the Phase 1 subgroup completing the retest at 14 days.

Ethical Considerations

This is a non-interventional observational study with minimal risk. No experimental treatments or invasive procedures are involved. All participants will be informed orally and in writing, and their non-opposition to participation will be documented before any data collection. The study complies with French regulations governing research involving human participants, data protection (GDPR), and Good Clinical Practice.

The sponsor of the study is Assistance Publique - Hôpitaux de Paris (AP-HP), with coordination by the Endocrinology, Diabetology and Nutrition Department at Ambroise Paré Hospital, and collaboration with the Neurosurgery Department at Lariboisière Hospital.