Vitamin D has been associated with depression, potentially via anti-inflammatory mechanisms, yet data are scarce, particularly in adolescence. We investigated (1) whether lower vitamin D status is associated with greater depression severity and (2) whether this association is statistically moderated by inflammation in patients of a child and adolescent psychiatry department. At admission, fasting morning venous blood was drawn. Serum vitamin D (25-hydroxy-cholecalciferol (25(OH)D)) and C-reactive protein (CRP) were analysed in all participants (n 465 (64·7 %♀; 11·3-18·9 years)). In a subsample (n 177), we additionally measured tumor necrosis factor-alpha (TNF-α), interferon-gamma (IFN-γ) and interleukin (IL)-1β, IL-6, IL-8 and IL-10. Depression severity was assessed by the Beck Depression Inventory II (BDI-II) (n 450), the Diagnostic System for Mental Disorders in Childhood and Adolescence via self-assessment (DISYPS Self) (n 441) and parent-assessment (DISYPS Proxy) (n 422). Overall, 43·2 % (n 201) were at risk for vitamin D deficiency (< 30 nmol/l), and 73·5-83·2 % - depending on assessment tool - showed at least mild depression. Linear regression revealed an inverse association between 25(OH)D and BDI-II in both crude and CRP-adjusted full-sample models. Logistic regressions showed a robust inverse association between 25(OH)D and DISYPS Proxy, but not for DISYPS Self. Although 25(OH)D was inversely correlated with some pro-inflammatory markers, neither their inclusion in regression models nor formal mediation analyses supported inflammation as a mediator of the vitamin D-depression association. Overall, our results suggest that vitamin D relates modestly to both depression and inflammation in adolescence. However, based on the measured parameters, we cannot confirm that anti-inflammatory effects are the link between vitamin D and depression.

The relationship between leptin levels and psychiatric disorders has been studied more extensively in adults than in children and adolescents. However, the results are conflicting. We investigated serum leptin levels in children and adolescents (11 to 18.9 years) with psychiatric disorders (n = 363). Absolute and relative (body-mass-index (BMI)-, sex- and pubertal-stage-adjusted z-scores using reference values of healthy children and adolescents) leptin levels of different patient groups according to diagnosis were compared. The association between leptin levels and depression (Beck Depression Inventory-II) and anxiety (Child Behavior Checklist and Youth Self Report) was examined using regression analysis. Leptin z-scores were higher in patients with psychiatric disorders than in healthy controls (median 1.50, p < .001). While global tests suggested differences in leptin z-scores between patients with different psychiatric disorders, these differences could not be attributed to diagnosis groups in post-hoc pairwise comparisons. Absolute leptin levels differed between psychiatric disorders (p < .001). Patients with anorexia nervosa (AN) had the lowest levels, and patients with mood disorders had higher leptin levels than patients with mental disorders other than mood disorders, anxiety or AN. Neither absolute nor relative leptin levels were related to depressive or anxiety symptoms in regression models adjusted for sex and BMI. Significantly elevated BMI-, sex- and puberty-stage-adjusted leptin levels were observed in children and adolescents with psychiatric disorders compared to a reference sample. Further controlled studies are needed to confirm and explain this finding. No relationship was found between absolute or relative leptin levels and symptoms of depression or anxiety.

Studies in adults suggest antidepressant effects of vitamin D, possibly via anti-inflammatory pathways, but evidence in youth is lacking. In the first randomized controlled trial (RCT) in depressed, vitamin D-deficient adolescents (DRKS00009758), symptom reduction emerged in only one of three outcomes. This secondary analysis investigates whether baseline inflammatory status modifies the effect of vitamin D. Data from 92 participants (78.3

Off-label treatment of a 17-year-old female patient with anorexia nervosa (AN), major depressive disorder (MDD), obsessive-compulsive disorder (OCD), posttraumatic stress disorder (PTSD), and non-suicidal self-injury (NSSI) with human recombinant leptin (metreleptin) for two dosing periods of 15 and seven days was associated with self- and clinician-rated improvements of eating disorder related psychopathology, OCD, depression, and NSSI. These intermittent improvements occurred despite the patient having high endogenous serum leptin levels ranging up to the 99th centile adjusted for sex, Tanner stage and body mass index (BMI; kg/m²) upon initiation of dosing. The results further extend the hypothesized effects of metreleptin on AN and comorbid psychopathology to patients with adjusted leptin levels in the normal to high range. We hypothesize that for unknown reasons the clinical symptomatology of a subgroup of patients with AN persists despite attainment of endogenous leptin levels well above those characteristic of the state of starvation, potentially entailing a prolonged entrapment in the eating disorder. An insight into the mechanisms underlying this distinct type of leptin resistance may help to promote recovery of such affected patients. Placebo controlled randomized clinical trials are required to assess if and to what extent short and medium term metreleptin treatment indeed improves recovery.

Purpose While observational studies revealed inverse associations between serum vitamin D levels [25(OH)D] and depression, randomized controlled trials (RCT) in children and adolescents are lacking. This RCT examined the effect of an untreated vitamin D deficiency compared to an immediate vitamin D 3 supplementation on depression scores in children and adolescents during standard day and in-patient psychiatric treatment. Methods Patients with vitamin D deficiency [25(OH)D ≤ 30 nmol/l] and at least mild depression [Beck Depression Inventory II (BDI-II) > 13] ( n = 113) were 1:1 randomized into verum (VG; 2640 IU vitamin D 3 /d) or placebo group (PG) in a double-blind manner. During the intervention period of 28 days, both groups additionally received treatment as usual. BDI-II scores were assessed as primary outcome, DISYPS-II (Diagnostic System for Mental Disorders in Childhood and Adolescence, Self- and Parent Rating) and serum total 25(OH)D were secondary outcomes. Results At admission, 49.3% of the screened patients ( n = 280) had vitamin D deficiency. Although the intervention led to a higher increase of 25(OH)D levels in the VG than in the PG (treatment difference: + 14 ng/ml; 95% CI 4.86–23.77; p = 0.003), the change in BDI-II scores did not differ (+ 1.3; 95% CI − 2.22 to 4.81; p = 0.466). In contrast, DISYPS parental ratings revealed pronounced improvements of depressive symptoms in the VG (− 0.68; 95% CI − 1.23 to − 0.13; p = 0.016). Conclusion Whereas this study failed to show a vitamin D supplementation effect on self-rated depression in adolescent in- or daycare patients, parents reported less depressive symptoms in VG at the end of our study. Future trials should consider clinician-rated depressive symptoms as primary outcome. Trial registration “German Clinical Trials Register” ( https://www.drks.de ), registration number: DRKS00009758

Based on the recent observation that human recombinant leptin (r-Met-hu-leptin; metreleptin) may induce a profound alleviation of the complex symptomatology of patients with anorexia nervosa (AN), we examine the implications for our conceptualisation of this eating disorder. Hypoleptinemia as a core endocrine feature of AN serves as a central and peripheral trigger of tissue-specific adaptations to starvation. In this narrative review, we argue that leptin deficiency may explain many of the puzzling features of this eating disorder. Weight loss can be viewed as a two-step process, with only the second step entailing hypoleptinemia and thereby the entrapment characteristic of AN. We discuss the central and peripheral distribution of leptin receptors and consider possible functional implications of hypoleptinemia. We contrast the slow psychological recovery of patients with AN and of people who experienced starvation upon weight recovery with the rapid onset of improvements upon off-label metreleptin treatment. Characteristics of the sex and age dependent secretion of leptin may contribute to the elevated vulnerability of young females to develop AN.

Abstract Both those with extreme obesity and anorexia nervosa (AN) are entrapped in their eating behavior and the associated cognitions and emotions. An overarching approach spanning these two disorders at the opposite ends of the distribution of the body mass index serves to highlight the necessity to search for potential endocrine or other endogenous mediators of both addictive-like overeating and undereating. The successful treatment of congenital leptin deficiency (CLD) with human recombinant leptin serves as a reminder that addictive-like overeating can have a biological basis. Similarly, rodent models for AN and promising case studies of patients with this eating disorder again point to a potential therapeutic role of human recombinant leptin. These encouraging results may help to transcend the current debate contrasting substance with non-substance–based addictive-like overeating.

Mutations leading to a reduced or loss of function in genes of the leptin-melanocortin system confer a risk for monogenic forms of obesity. Yet, gain of function variants in the melanocortin-4-receptor ( MC4R ) gene predispose to a lower BMI. In individuals with reduced body weight, we thus expected mutations leading to an enhanced function in the respective genes, like leptin ( LEP) and MC4R . Therefore, we have Sanger sequenced the coding regions of LEP and MC4R in 462 female patients with anorexia nervosa (AN), and 445 healthy-lean controls. In total, we have observed four and eight variants in LEP and MC4R , respectively. Previous studies showed different functional in vitro effects for the detected frameshift and non-synonymous variants: (1) LEP : reduced/loss of function (p.Val94Met), (2) MC4R : gain of function (p.Val103Ile, p.Ile251Leu), reduced or loss of function (p.Thr112Met, p.Ser127Leu, p.Leu211 fs X) and without functional in vitro data (p.Val50Leut). In LEP , the variant p.Val94Met was detected in one patient with AN. For MC4R variants, one patient with AN carried the frameshift variant p.Leu211 fs X. One patient with AN was heterozygous for two variants at the MC4R (p.Val103Ile and p.Ser127Leu). All other functionally relevant variants were detected in similar frequencies in patients with AN and lean individuals.

ABSTRACTObjectivePost‐operative development of restrictive eating disorders can occur in patients after bariatric surgery. In children and adolescents with anorexia nervosa (AN) or atypical AN, premorbid body mass index (BMI) has recently been shown to predict total weight loss. We hypothesized that pre‐operative BMI similarly predicts weight loss and the development of a restrictive eating disorder in adult bariatric patients.MethodA PubMed search identified case studies/series of 29 adult females who developed AN or atypical AN/eating disorder not otherwise specified following bariatric surgery. Non‐parametric Spearman's correlation (r s ) between pre‐operative BMI and total weight loss was calculated; a scatterplot was used to illustrate the relationship between pre‐operative/premorbid BMI and weight loss in kg for 29 bariatric patients and 460 children and adolescents with AN or atypical AN as published previously.ResultsThe correlation between pre‐operative BMI and weight loss among bariatric patients was r s = 0.65 (p = 0.0001). Scatterplot data of this relationship fit the previously identified pattern in children and adolescents with AN or atypical AN.DiscussionThe prediction of weight loss by pre‐operative/premorbid BMI appears applicable across the weight spectrum, from underweight to severe obesity, thus strengthening our hypothesis of underlying regulatory mechanisms for the development of AN and atypical AN. Such data may guide the determination of critical weight loss thresholds that trigger eating disorder development in predisposed individuals.

Off-label metreleptin treatment resulted in cognitive, emotional and behavioral improvements of patients with anorexia nervosa, who presented with hypoleptinemia. We now report a case study of a 16-year-old female patient with atypical anorexia nervosa who was treated off-label with metreleptin for 11 days. She had lost 21 kg over 6 months. Her body mass index at referral for inpatient treatment was 20 kg/m 2 , her serum leptin level was just within the normal range (2.4 ng/ml). Dosing resulted in prominent improvements of mood and weight phobia entailing a comparatively brief inpatient treatment. The observed improvements are similar to those observed in patients with AN, suggesting overlapping mechanisms with respect to clinical effects induced by elevations of absolute or relative hypoleptinemia. Randomized controlled trials are warranted for both eating disorders.

Case studies of patients with anorexia nervosa revealed psychological improvements upon treatment with recombinant human leptin, including an alleviation of the entrapment in this eating disorder. Hypoleptinaemia induced by loss of adipose tissue mass might trigger psychological changes observed in starvation. Accordingly, leptin analogues might prove beneficial for treatment of anorexia nervosa.

A comorbidity of anorexia nervosa (AN) and myalgic encephalomyelitis (ME/CSF) is uncommon. A 17 years-old male adolescent with possible onset of ME/CFS after an Epstein Barr Virus infection (EBV) and later onset of AN during a second period of weight loss was twice treated off-label with metreleptin for 15 and 11 days, respectively. As in previous cases, eating disorder specific cognitions and mood improved. Interestingly, fatigue and post-exertional muscle pain (P-EMP) improved, too. We discuss potential mechanisms. Treatment with metreleptin may prove beneficial in AN and in ME/CSF associated with substantial weight loss.

This narrative review addressed to both clinicians and researchers aims to assess the role of hypoleptinemia in disordered sleep with a particular focus on patients with anorexia nervosa (AN). After introducing circadian rhythms and the regulation of circulating leptin, we summarize the literature on disordered sleep in patients with AN and in fasting subjects in general. We highlight novel single-case reports of substantially improved sleep within days after initiation of off-label metreleptin treatment. These beneficial effects are set in relationship to current knowledge of disordered sleep in animal models of an impaired leptin signaling. Specifically, both absolute and relative hypoleptinemia play a major role in animal models for insomnia, obstructive sleep apnea and obesity hypoventilation syndrome. We pinpoint future research required to complement our understanding of the role of leptin in sleep in patients with acute AN. Moreover, within the section clinical applications we speculate that human recombinant leptin may be useful for the treatment of treatment-resistant sleep-wake disorders, which are associated with (relative) hypoleptinemia. Overall, we stress the role of the hormone leptin in sleep.

Off-label treatment of a 15-year-old female patient with anorexia nervosa (AN) with human recombinant leptin (metreleptin) for nine days was associated with self-reported increments of appetite and hunger resulting in rapid weight gain and substantial improvement of eating disorder cognitions and of depression. The results further substantiate the effects of metreleptin on both AN and depression. We contrast these results with the widespread view that leptin is an anorexigenic hormone. Randomized controlled trials are warranted to confirm the described effects.

Zusammenfassung. Genetische Varianten beeinflussen die Gewichtsregulation und die Entwicklung von Essstörungen. Zunächst haben familienbasierte, sogenannte formalgenetische Studien den erblichen Anteil an der Gewichtsregulation und an der Ätiologie von Essstörungen beleuchtet. In einer Vielzahl von Studien zeigten sich sowohl für die Varianz des Körpergewichts als auch für die Entstehung von Essstörungen Erblichkeitsschätzer (Heritabilitätsraten) von über 50 %. Mit diesem Wissen begab man sich in den 90er-Jahren des letzten Jahrhunderts auf die Suche nach den zugrundeliegenden Genen (genauer: genetischen Varianten), die das Körpergewicht, das Essverhalten oder beide Phänotypen auf Grundlage geteilter Mechanismen beeinflussen. Zunächst wurden Kandidatengenstudien durchgeführt. Dabei untersuchte man auf Grundlage unterschiedlicher, v. a. aber pathophysiologisch plausibler Überlegungen Gene mit hoher Relevanz für die untersuchten Phänotypen. Dieser Ansatz war für Essstörungen nicht sehr erfolgreich, für die Gewichtsregulation konnte eine Handvoll Gene identifiziert werden. Verbunden mit großen methodischen Fortschritten in der genetischen Forschung und v. a. der Etablierung sogenannter genomweiter Assoziationsstudien (GWAS) Anfang der 2000er-Jahre konnten bislang über 1000 Varianten/Genorte detektiert werden, die das Körpergewicht beeinflussen. Für die Essstörung Anorexia nervosa (AN) sind aktuell acht solcher Genorte beschrieben. Diese Ergebnisse, aber auch aktuelle Ansätze zu phänotypübergreifenden Analysen lassen Einblicke in die komplexe Regulation des Körpergewichtes zu und haben zudem unerwartete Pathomechanismen für AN aufgezeigt.

Two psychiatric interviews of a 39-year old female with congenital leptin deficiency were conducted to define psychological changes fourteen and 165 days after initiation of treatment with human recombinant leptin (metreleptin). The most pronounced initial experience related to the reduced preoccupation with food. An improved mood was reported by the patient, which she associated with this reduced preoccupation. Her mood remained elevated upon recontact, whereas she was no longer preoccupied with food. Overall, the interviews provides a vivid account of the subjective experiences upon initiation of treatment. Some of the findings bear resemblance to those reported recently in patients with anorexia nervosa who were treated with metreleptin for one to three weeks. This case report provides further evidence that metreleptin has strong psychophamacological effects in patients with absolute or relative leptin deficiency. We strongly recommend profound psychological examinations of patients with congenital leptin deficiency at baseline and after intitiation of treatment with human recombinant leptin to gain further insight into the functions affected by this hormone.